Document 11612149

advertisement
AN ABSTRACT OF THE THESIS OF
Freddy Mauricio Lemus for the degree of Master of Science in Food Science and
Technology presented on September 19, 2012
Title: Flavor Development of Cheddar Cheese Under Different Manufacturing
Practices
Abstract approved:
Michael C. Qian
Cheddar Cheese samples (good cheese, weak cheese, cheese made with pasteurized milk,
cheese made with heat-shocked milk, cheese from production plant A, cheese from production
plant B, cheese made with adjunct culture, and cheese made without adjunct culture), were
evaluated during the ripening stage. Proteolysis was studied by a fractionation scheme,
resulting in an insoluble fraction analyzed by urea polyacrylamide gel electrophoresis (UreaPAGE), and a soluble fraction which was further investigated through water soluble nitrogen
(WSN), trichloroacetic acid soluble nitrogen (TCA-SN) and phosphotungstic acid soluble
nitrogen (PTA-SN) analyzed by total Kjeldahl nitrogen content (TKN). Reversed phase high
performance liquid chromatography (RP-HPLC) was used to study the peptide profile of the
water soluble fraction. Lipolyisis was studied by levels of individual free fatty acids
determined through gas chromatography-flame ionization detection (GC-FID) after isolation
employing solid phase extraction (SPE). Volatile sulfur compounds were studied using head
space solid phase micro-extraction (SPME) coupled with gas chromatography-pulsed flame
photometric detection (PFPD).
It was found that Urea-PAGE is capable to differentiate samples according their age, but
cannot discriminate samples regarding the treatment assessed, quality or origin of the samples.
However, measurements of total Kjeldahl Nitrogen (TKN) of the WSN, TCA-SN, and PTA-
SN fractions, and the principal component analysis of the RP-HPLC peptide profile of the
WSN fraction, revealed differences in the rate and pattern of proteolysis for each one of the
manufacturing cases. Good cheese, cheese produce in plant TCCA, cheese made in plant CRP
with adjunct culture isolated from plant TCCA cheese, and cheese made with heat-shocked
milk developed higher level of total nitrogen for the WSN, TCA-SN and PTA-SN fractions,
indicating that primary and secondary proteolysis were faster for these samples. This is
supported by a PCA model with three principal components that account for the 80-83% of the
variability of the data from the RP-HPLC peptide profile analysis, which discriminates the
samples according to age and manufacturing practice. In addition, FFA profiles demonstrated
higher levels of low and medium chain free fatty acids for good cheese, cheese produce in
plant TCCA, cheese made in plant CRP with adjunct culture, and cheese made with heatshocked milk samples, which suggest faster lipolysis during ripening. The Volatile Sulfur
Compounds (VSC) analysis showed higher levels of DMS and MeSH and lower levels of H2S,
suggesting faster catabolism of sulfur containing amino acids in good cheese, cheese produce
in plant TCCA, cheese made in plant CRP with adjunct culture, and cheese made with heatshocked milk.
© Copyright by Freddy Mauricio Lemus
September 19, 2012
All Rights Reserved
Flavor Development of Cheddar Cheese Under Different Manufacturing Practices
by
Freddy Mauricio Lemus
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Presented September 19, 2012
Commencement June 2013
Master of Science thesis of Freddy Mauricio Lemus
presented on September 19, 2012
APPROVED:
Major Professor, representing Food Science and Technology
Head of the Department of Food Science and Technology
Dean of the Graduate School
I understand that my thesis will become part of the permanent collection of Oregon
State University libraries. My signature below authorizes the release of my thesis to
any reader request
Freddy Mauricio Lemus Munoz, Author
ACKNOWLEDGEMENTS
I sincerely thank Dr. Michael C. Qian, for his invaluable support, guidance,
intellectual stimulation and especially for his patience and trust during the most
challenging and difficult period of my life and my days at Oregon State University.
I thank Dr. Robert Mc Gorrin, Head of the department of Food Science and
Technology, Dr. Lisbeth Goddik, Dr. Thomas Shellhammer and Dr. Manoj
Pastey, members of my Graduate Committee for providing an opportunity to pursue
the M.S in Food Science and Technology at Oregon State University and for their
continuous willing to help.
My special thanks go to Cody Osborn and Family, Daniel Sharp and Family,
Phillip Widstock, Helena Sanchez, Barb Kralj, Salaheddine Ziadeh, Nancy and
Faith for their sincere and unconditional friendship that gave me happiness, keep me a
float and was light during the darkest moments, and without whom this would not be
possible.
I would like to thank Dr. Yan Ping Qian, for her constant support, significant inputs,
and absolute and unreserved collaboration during the various stages of this research
project.
I certainly would like to thank Linda Dunn and Linda Hoyser for their assistance to
come here and for helping me to make a dream reality.
I would like to thank the Office Staff for their wonderful work
I appreciate the enthusiasm, strength and commitment of Hui Fang, Qin Zou and
Grace, whose assistance and inputs were fundamental for my research work.
Finally and most importantly, I thank my family... mi viejita la luna rockera, el Gordis,
Sarah, Jan… and the best and only one … mi calvo, mi parce, mi padre y mi hermano,
mi Heverth… for all of you there are no words enough!!
CONTRIBUTION OF AUTHORS
Dr. Michael C Qian, my advisor and pillar of this project, arranging the required
funding, supervision and assistance at every stage of the project. Dr Will Austin
provided some of the facilities of the Central analytical Laboratory for analysis of
plant tissue, soil and water for nutrients at Oregon State University. Dr Yan Ping
Qian, assisted the development of methodology for the Urea PAGE electrophoresis
and TKN Kjeldahl analyses. Hui Fang assisted in the collection of some FFA analysis
data. And Helen Burbank assisted in the collection of some VSC analysis data.
TABLE OF CONTENTS
Page
CHAPTER 1----------------------------------------------------------------------------------------------------- 1
INTRODUCTION --------------------------------------------------------------------------------------------- 1
BACKGROUND ........................................................................................................................ 2
LITERATURE REVIEW ........................................................................................................... 3
CHEDDAR CHEESE ............................................................................................................. 3
FLAVOR FORMATION OF CHEESE ................................................................................. 3
Glycolysis ........................................................................................................................... 4
Lipolysis ............................................................................................................................. 4
Proteolysis .......................................................................................................................... 7
FACTORS AFFECTING THE DEVELOPMENT OF VOLATILE COMPOUNDS ......... 11
Milk preparation: selection, pasteurization & standardization ......................................... 12
Acidification ..................................................................................................................... 13
Coagulation....................................................................................................................... 13
Syneresis ........................................................................................................................... 14
Cheddaring........................................................................................................................ 15
Salting ............................................................................................................................... 15
Pressing and molding........................................................................................................ 16
CHEESE FLAVOR AND FLAVOR ANALYSIS ............................................................... 17
Instrumental analysis of cheddar cheese........................................................................... 19
Gas Chromatography .................................................................................................... 20
Extraction-Distillation .............................................................................................. 20
Headspace methods .................................................................................................. 21
Separation ................................................................................................................. 23
Reverse Phase-High Performance Liquid Chromatography (RP-HPLC) ..................... 24
Electrophoresis ............................................................................................................. 26
Aroma of Cheddar cheese................................................................................................. 28
REFERENCES ......................................................................................................................... 31
CHAPTER 2--------------------------------------------------------------------------------------------------- 44
TABLE OF CONTENTS (CONTINUED)
EVALUATION OF THE EFFECT OF HEAT-SHOCK AND PASTEURIZATION
TREATMENTS OF CHEESE MILK ON RIPENING RELATED CHANGES IN CHEDDAR
CHEESE ------------------------------------------------------------------------------------------------------- 44
ABSTRACT ............................................................................................................................. 45
INTRODUCTION .................................................................................................................... 46
MATERIALS AND METHODS ............................................................................................. 48
CHEESE SAMPLES ............................................................................................................ 48
FREE FATTY ACIDS ANALYSIS ..................................................................................... 48
Chemicals ......................................................................................................................... 48
Extraction.......................................................................................................................... 49
Chromatography ............................................................................................................... 49
Quantitative analysis......................................................................................................... 49
VOLATILE SULFUR COMPOUNDS (VSC’S) ................................................................. 50
Chemicals ......................................................................................................................... 50
Extraction.......................................................................................................................... 50
Chromatography ............................................................................................................... 50
Quantitative analysis......................................................................................................... 51
Matrix effect ................................................................................................................. 51
Sulfur standards and internal standard preparation ....................................................... 51
PROTEOLYSIS .................................................................................................................... 52
Chemicals ......................................................................................................................... 52
Sample preparation and fractionation ............................................................................... 52
Macro blog digestion (Kjeldahl Digestion) ...................................................................... 53
Reversed phase High performance liquid chromatography analysis ................................ 53
Electrophoresis ................................................................................................................. 54
STATISTICAL ANALYSIS ................................................................................................ 54
RESULTS AND DICUSSION ................................................................................................. 54
FREE FATTY ACIDS FFA ................................................................................................. 54
Volatile Sulfur Compounds VSC’s....................................................................................... 58
Effect of treatment on Proteolysis ........................................................................................ 63
Soluble Nitrogen Fractions and TKN ............................................................................... 63
Peptide analysis by RP-HPLC .......................................................................................... 68
Electrophoresis ................................................................................................................. 75
CONCLUSION ........................................................................................................................ 77
REFERENCES ......................................................................................................................... 77
TABLE OF CONTENTS (CONTINUED)
CHAPTER 3--------------------------------------------------------------------------------------------------- 90
COMPARISON OF RIPENING CHARACTERISTICS OF CHEDDAR CHEESE FROM
THE SAME COMPANY MANUFACTURED IN DIFFERENT PRODUCTION PLANTS 90
ABSTRACT ............................................................................................................................. 91
INTRODUCTION .................................................................................................................... 92
MATERIALS AND METHODS ............................................................................................. 94
CHEESE SAMPLES ............................................................................................................ 94
FREE FATTY ACIDS ANALYSIS ..................................................................................... 94
Chemicals ......................................................................................................................... 94
Extraction.......................................................................................................................... 94
Chromatography ............................................................................................................... 95
Quantitative analysis......................................................................................................... 95
VOLATILE SULFUR COMPOUNDS (VSC’S) ................................................................. 95
Chemicals ......................................................................................................................... 95
Extraction.......................................................................................................................... 96
Chromatography ............................................................................................................... 96
Quantitative analysis......................................................................................................... 96
Matrix effect ................................................................................................................. 96
Sulfur standards and internal standard preparation ....................................................... 97
PROTEOLYSIS .................................................................................................................... 98
Chemicals ......................................................................................................................... 98
Sample preparation and fractionation ............................................................................... 98
Macro blog digestion (Kjeldahl Digestion) ...................................................................... 98
Reversed phase High performance liquid chromatography analysis ................................ 99
Electrophoresis ................................................................................................................. 99
STATISTICAL ANALYSIS .............................................................................................. 100
RESULTS AND DICUSSION ............................................................................................... 100
FREE FATTY ACIDS FFA ............................................................................................... 100
Volatile Sulfur Compounds VSC’s..................................................................................... 103
Effect of treatment on Proteolysis ...................................................................................... 108
Soluble Nitrogen Fractions and TKN ............................................................................. 108
Peptide analysis by RP-HPLC ........................................................................................ 112
Electrophoresis ............................................................................................................... 118
CONCLUSION ...................................................................................................................... 120
TABLE OF CONTENTS (CONTINUED)
REFERENCES ....................................................................................................................... 121
CHAPTER 4-------------------------------------------------------------------------------------------------- 126
COMPARISION OF CHEDDAR CHEESE RIPENING MANUFACTURED WITH AND
WITHOUT ADJUNCT CULTURE --------------------------------------------------------------------- 126
ABSTRACT ........................................................................................................................... 127
INTRODUCTION .................................................................................................................. 128
MATERIALS AND METHODS ........................................................................................... 130
CHEESE SAMPLES .......................................................................................................... 130
FREE FATTY ACIDS ANALYSIS ................................................................................... 130
Chemicals ....................................................................................................................... 130
Extraction........................................................................................................................ 130
Chromatography ............................................................................................................. 131
Quantitative analysis....................................................................................................... 131
VOLATILE SULFUR COMPOUNDS (VSC’S) ............................................................... 131
Chemicals ....................................................................................................................... 131
Extraction........................................................................................................................ 132
Chromatography ............................................................................................................. 132
Quantitative analysis....................................................................................................... 133
Matrix effect ............................................................................................................... 133
Sulfur standards and internal standard preparation ..................................................... 133
PROTEOLYSIS .................................................................................................................. 134
Chemicals ....................................................................................................................... 134
Sample preparation and fractionation ............................................................................. 134
Macro blog digestion (Kjeldahl Digestion) .................................................................... 134
Reversed phase High performance liquid chromatography analysis .............................. 135
Electrophoresis ............................................................................................................... 135
STATISTICAL ANALYSIS .............................................................................................. 136
RESULTS AND DICUSSION ............................................................................................... 136
FREE FATTY ACIDS FFA ............................................................................................... 136
Volatile Sulfur Compounds VSC’s..................................................................................... 139
Effect of treatment on Proteolysis ...................................................................................... 142
Soluble Nitrogen Fractions and TKN ............................................................................. 142
Peptide analysis by RP-HPLC ........................................................................................ 146
Electrophoresis ............................................................................................................... 152
TABLE OF CONTENTS (CONTINUED)
CONCLUSION ...................................................................................................................... 154
REFERENCES ....................................................................................................................... 155
CHAPTER 5-------------------------------------------------------------------------------------------------- 159
COMPARATIVE CHEMICAL ANALYSIS OF GOOD AND WEAK CHEDDAR CHEESE
DURING RIPENING -------------------------------------------------------------------------------------- 159
ABSTRACT ........................................................................................................................... 160
INTRODUCTION .................................................................................................................. 161
MATERIALS AND METHODS ........................................................................................... 162
CHEESE SAMPLES .......................................................................................................... 162
FREE FATTY ACIDS ANALYSIS ................................................................................... 162
Chemicals ....................................................................................................................... 162
Extraction........................................................................................................................ 162
Chromatography ............................................................................................................. 163
Quantitative analysis....................................................................................................... 163
VOLATILE SULFUR COMPOUNDS (VSC’S) ............................................................... 163
Chemicals ....................................................................................................................... 163
Extraction........................................................................................................................ 164
Chromatography ............................................................................................................. 164
Quantitative analysis....................................................................................................... 165
Matrix effect ............................................................................................................... 165
Sulfur standards and internal standard preparation ..................................................... 165
PROTEOLYSIS .................................................................................................................. 166
Chemicals ....................................................................................................................... 166
Sample preparation and fractionation ............................................................................. 166
Macro blog digestion (Kjeldahl Digestion) .................................................................... 167
Reversed phase High performance liquid chromatography analysis .............................. 167
Electrophoresis ............................................................................................................... 167
STATISTICAL ANALYSIS .............................................................................................. 168
RESULTS AND DICUSSION ............................................................................................... 168
FREE FATTY ACIDS FFA ............................................................................................... 168
Volatile Sulfur Compounds VSC’s..................................................................................... 172
Effect of treatment on Proteolysis ...................................................................................... 177
Soluble Nitrogen Fractions and TKN ............................................................................. 178
TABLE OF CONTENTS (CONTINUED)
Peptide analysis by RP-HPLC ........................................................................................ 182
Electrophoresis ............................................................................................................... 188
CONCLUSION ...................................................................................................................... 189
REFERENCES ....................................................................................................................... 190
CONCLUSION ---------------------------------------------------------------------------------------------- 195
CONCLUSION ---------------------------------------------------------------------------------------------- 195
BIBLIOGRAFY --------------------------------------------------------------------------------------------- 197
APPENDICES ----------------------------------------------------------------------------------------------- 209
APENDIX A ........................................................................................................................... 209
ANOVAS ............................................................................................................................... 209
FFA ..................................................................................................................................... 209
Heat-shocked Vs Pasteurized ......................................................................................... 209
TCCA Vs CRP................................................................................................................ 211
Adjunct culture Vs Not ................................................................................................... 213
Good Vs Week................................................................................................................ 216
VSC’s.................................................................................................................................. 218
Heat-shocked Vs Pasteurized ......................................................................................... 218
TCCA Vs CRP................................................................................................................ 219
Adjunct culture Vs Not ................................................................................................... 219
Good Vs Weak................................................................................................................ 220
TKN .................................................................................................................................... 221
Heat-shocked Vs Pasteurized ......................................................................................... 221
TCCA Vs CRP................................................................................................................ 222
Adjunct culture Vs Not ................................................................................................... 222
Good Vs Weak................................................................................................................ 223
APENDIX B ........................................................................................................................... 225
PCA CORRELATION MATRIX .......................................................................................... 225
HEAT SHCOKED VS PASTEURIZATION ..................................................................... 225
TCCA Vs CRP.................................................................................................................... 225
ADJUCNT CULTURE Vs NOT ........................................................................................ 226
GOOD VS WEAK............................................................................................................. 227
LIST OF FIGURES
Figure 1 FFA chromatogram Heat –schokecd Vs Pasteurized milk ......................................... 56
Figure 2 Development of individual FFA in Heat-shocked and Pasteurized milk cheese ....... 56
Figure 3 VSC's chromatogram Heat –schocked Vs Pasteurized milk ...................................... 62
Figure 4 Calibration curve DMS .............................................................................................. 62
Figure 5 Calibration Curve H2S ................................................................................................ 62
Figure 6 Development of MeSH in Heat-schocked and Pasteurized milk ............................... 62
Figure 7 Development of H2S in Heat-schocked and Pasteurized milk ................................... 62
Figure 8 Development of DMS in Heat-schocked and Pasteurized milk ................................. 63
Figure 9 TKN fractions heat-schocked cheese milk ................................................................. 67
Figure 10 TKN fractions pasteurized cheese milk.................................................................... 67
Figure 11 WSN heat-shocked and pasteurized cheese milk ..................................................... 67
Figure 12 TCA heat-schocked and pasteurized cheese milk .................................................... 68
Figure 13 PTA-SN heat-schocked and pasteurized cheese milk .............................................. 68
Figure 14 Scree plot of heat-shocked and pasteurized cheese milk.......................................... 73
Figure 15 Score plot of heat-shocked and pasteurized cheese milk (age) ................................ 73
Figure 16 Score plot of heat-shocked and pasteurized cheese milk (temperature) ................... 74
Figure 17 Loading plot for heat-shocked and pasteurized cheese milk .................................... 74
Figure 18 Peptide profile pasteurized cheese (A and B) Vs Heat-shocked (C and D) milk, for 2
and 12 months ........................................................................................................................... 75
Figure 19 Urea PAGE for ripening of cheese made from heat-shocked milk .......................... 76
Figure 20 Urea PAGE for heat-schocked Vs pasteurized cheeses (12 months) ....................... 76
Figure 21 FFA chromatogram TCCA Vs CRP....................................................................... 101
Figure 22 Development of individual FFA for TCCA and CRP cheese ................................ 102
Figure 23 VSC's chromatogram for TCCA cheese ................................................................. 107
LIST OF FIGURES (CONTINUED)
Figure 24 Development of H2S for TCCA Vs CRP cheese .................................................... 107
Figure 25Development of MeSH for TCCA Vs CRP cheese ................................................. 107
Figure 26 Development of DMS for TCCA Vs CRP cheese ................................................. 108
Figure 27 TKN fractions TCCA cheese ................................................................................. 110
Figure 28 TKN fractions CRP cheese..................................................................................... 111
Figure 29 WSN TCCA Vs CRP cheese .................................................................................. 111
Figure 30 TCA-SN TCCA Vs CRP cheese ............................................................................ 111
Figure 31 PTA-SN TCCA Vs CRP cheese ............................................................................. 112
Figure 32 Scree plot of TCCA Vs CRP cheese ...................................................................... 116
Figure 33 Score plot of TCCA Vs CRP cheese (age) ............................................................. 116
Figure 34 Score plot of TCCA Vs CRP cheese (Origin) ........................................................ 117
Figure 35 Loading plot of TCCA Vs CRP cheese .................................................................. 117
Figure 36 Peptide profile CRP cheese (A and B) Vs TCCA (C and D) cheese, for 2 and 14
months .................................................................................................................................... 118
Figure 37 Urea PAGE for ripening of TCCA cheese ............................................................. 119
Figure 38 Urea PAGE for TCCA Vs CRP cheeses (12 m)..................................................... 120
Figure 39 Urea PAGE for TCCA Vs CRP cheeses (2 months) .............................................. 137
Figure 40 Development of individual FFA for cheese with and without adjunct culture....... 137
Figure 41 VSC chromatogram for Cheese with adjunct culture ............................................. 141
Figure 42 Development of H2S in Cheese made with and without adjunct culture................ 141
Figure 43 Development of MeSH in Cheese made with and without adjunct culture............ 141
Figure 44 Development of DMS in Cheese made with and without adjunct culture ............. 142
Figure 45 TKN fractions cheese with adjunct culture ........................................................... 144
Figure 46 TKN fractions cheese without adjunct culture ...................................................... 145
LIST OF FIGURES (CONTINUED)
Figure 47 WSN Cheese with Adj culture Vs Cheese without ................................................ 145
Figure 48 TCA-SN Cheese with Adj culture Vs Cheese without ........................................... 145
Figure 49 PTA-SN Cheese with Adj culture Vs Cheese without ........................................... 146
Figure 50 Scree plot of cheese with and without adjunct culture ........................................... 150
Figure 51 Score plot of cheese with and without adjunct culture (age).................................. 150
Figure 52 Score plot of cheese with and without adjunct culture (culture) ............................ 151
Figure 53 Loading plot of cheese with and without adjunct culture....................................... 151
Figure 54 Peptide profile cheese with adjunct culture (A and B) Vs Not (C and D) cheese, for
2 and 10 months ...................................................................................................................... 152
Figure 55 Urea PAGE for ripening of cheese made with and without adjunct culture .......... 153
Figure 56 Urea PAGE for ripening of cheese made with and without adjunct culture .......... 154
Figure 57 FFA chromatogram week cheese ........................................................................... 169
Figure 58 Development of individual FFA in good cheese vs weak cheese .......................... 170
Figure 59 VSC's chromatogram Good Cheese ....................................................................... 175
Figure 60 Calibration Curve H2S ............................................................................................ 176
Figure 61 Calibration Curve DMS ......................................................................................... 176
Figure 62 Development of MeSH in Good and weak cheese ................................................. 176
Figure 63 Development of H2S in Good and weak cheese .................................................... 177
Figure 64 Development of DMS in Good and weak cheese.................................................. 177
Figure 65 TKN fractions Good cheese ................................................................................... 180
Figure 66 TKN fractions Good cheese ................................................................................... 180
Figure 67 WSN Good Vs Weak cheese .................................................................................. 181
Figure 68 WSN Good Vs Weak cheese .................................................................................. 181
Figure 69 PTA-SN Good Vs Weak cheese............................................................................. 182
LIST OF FIGURES (CONTINUED)
Figure 70 Scree plot of Good Vs Weak cheese ...................................................................... 185
Figure 71 Score plot of Good Vs Weak cheese (age) ............................................................ 186
Figure 72 Score plot of Good Vs Weak cheese (quality) ....................................................... 186
Figure 73 Peptide profile for heat-shocked and pasteurized cheese milk ............................... 186
Figure 74 Peptide profile Weak Cheese (A and B) Vs Good Cheese (C and D) .................... 187
Figure 75 Urea PAGE for ripening of Good cheese ............................................................... 188
Figure 76 77 Urea PAGE for Good Vs Weak cheese (12 months) ........................................ 189
LIST OF TABLES
Table 1 Aroma compounds of Cheddar cheese ........................................................................ 28
Table 2 Anova FFA of Heat-shocked Vs Pasteurized ............................................................ 209
Table 3 Anova FFA of TCCA Vs CRP ................................................................................. 211
Table 4 Anova FFA of Adjunct culture Vs Not...................................................................... 213
Table 5 Anova FFA of Good Vs Weak .................................................................................. 216
Table 6 Anova VSC of Heat-Shocked Vs Pasteurized ........................................................... 218
Table 7 Anova VSC TCCA Vs CRP ...................................................................................... 219
Table 8 Anova of VSC Adjunct Vs Not ................................................................................. 219
Table 9 Anova of VSC Good Vs Weak .................................................................................. 220
Table 10 Anova of TKN Heat-Shocked Vs Pasteurized......................................................... 221
Table 11 TKN TCCA Vs CRP ............................................................................................... 222
Table 12 TKN Adjunct culture Vs Not ................................................................................... 222
Table 13 TKN Good Vs Weak ............................................................................................... 223
Table 14 Correlation matrix for HEAT SHCOKED VS PASTEURIZATION...................... 225
Table 15 Coreelation matrix of TCCA Vs CRP ..................................................................... 226
Table 16 Correlation matrix of Adj Vs Not ............................................................................ 226
Table 17 Correlation matrix Good Vs Weak .......................................................................... 227
DEDICATED TO MY BROTHER
FLAVOR DEVELOPMENT OF CHEDDAR CHEESE UNDER DIFFERENT
MANUFACTURING PRACTICES
CHAPTER 1
INTRODUCTION
2
BACKGROUND
The quality of food products is often associated to their flavor. It is essentially determined by
sensations and perceptions in our body resulting from metabolic responses to flavor
compounds in those foods we eat. These impressions are collected by the set of complex
detectors in our senses and scanned by our brains. Whereby, only through the presence of
certain sapid and volatile compounds, at specific levels and ratios, it is possible to determine
the characteristic odor and flavor of a product. Therefore, those compounds with greater
impact on the perceived aroma and taste, known as character impact compounds, are matter of
devoted research.
Cheddar cheese is a low temperature hard rennet-coagulated cheese, traditionally made from
cow’s milk, high in fat and solids, resulting in a cheese with firm consistency, with no holes
and a flavor described as mild or pungent depending on age. The main operations during
cheese manufacturing that induce flavor formation include milk selection, standardization and
heat treatment, pre-acidification, addition of starter and adjunct cultures, coagulation, cooking,
washing of curds, determination of size and shape of cheese, pressing, resting, salting, and
ripening of curds. Other factors affecting flavor of Cheddar cheese include salt in moisture
level and fat content (P. Walstra et al. 1999)
The aim of this work was to create a model to assess quality variables during the maturation of
the cheddar cheese system, which could be used to construct a set of data to propose a finger
print of quality parameter to reproduce through the proper handling of processing variables.
3
LITERATURE REVIEW
CHEDDAR CHEESE
Cheddar cheese is a fermented milk base food product, resulting from a two stages
dehydration process: 1) curds preparation and 2) ripening of curds; where fats and caseins are
concentrated 6 to 12 fold. It is as well a low temperature, hard, and unwashed variety,
traditionally made using mesophilic starter culture.
The US Food and Drug Administration (FDA) states that cheddar cheese may be classified as
a food product prepared by any procedure resulting in a cheese with a minimum milk fat
content of 50% by weight of solids, and a maximum moisture content of 39% by weight (FDA
2011). Its moisture content and water activity (aw) vary from 30 to 50% and from 0.87 to 0.98
respectively, and has a pH between 5.0 and 5.3 (P. Walstra et al. 1999). Other typical
composition by weight percentage for cheddar cheese is: protein (24.9%), fat (33%), total
CHO (1.3%), Ash (3.9%), Ca (0.72%), P (0.51%), salt (1.8%), and salt in moisture (4.9%).
In addition, about 10.6 billion pounds of cheese were produced in the United States in 2011,
and around the 33% of the total production was Cheddar cheese (NASS, 2011).
FLAVOR FORMATION OF CHEESE
Flavor compounds in cheese arise from biochemical reactions happening during the ripening
stage. These are mainly the degradation of proteins (caseins), lipids, lactose and citrate in
milk, and subsequent catabolic reactions. They are grouped into:

Glycolysis: metabolism of lactose and citrate

Lipolysis: liberation of free fatty acids (FFA) from triacylglycerols, and
subsequent metabolism to volatile compounds

Proteolysis: degradation of casein matrix into peptides, and ultimately free amino
acids (FAA), followed by the catabolism of FAA to produce flavor compounds,
such as carboxylic acids and sulfur compounds.
4
Glycolysis
Although lactose metabolism does not contribute directly to cheese flavor, it is significant at
determining the texture of the cheese and therefore the rate of liberation of sapid compound.
During the syneresis stage most of the lactose is lost in the whey drainage. However, the 0.8 1.5% of the total lactose remains in the cheese curd as substrate for the starter lactic acid
bacteria (LAB) (Huffman and Kristoffersen 1984). In the case of Cheddar cheese, it is mostly
fermented to lactic acid L+ before salting and molding (P. F. Fox, Lucey, and Cogan 1990),
because LAB cultures are stimulated by low levels of NaCl and are strongly inhibited over
2.5% (Turner and Thomas 1980). Whereby, salting controls the metabolism of lactose by
decreasing the starter activity and lactic acid production.
When LAB activity has declined, shortly after the addition of salt, nonstarter lactic acid
bacteria (NLAB), metabolizes the remaining lactose to DL-lactate and then racemizes it to Llactate (P. F. Fox, Lucey, and Cogan 1990). This last reaction is not significant for the flavor
of cheese, but is determining for its structure and texture, since racemization support the
development of calcium-D Lactate crystals, which are incorrectly interpreted by consumers as
spoilage, resulting in product rejection. In addition lactate can be metabolized by LAB and
NLAB to CO2 and acetate, which at high concentration may be perceive as an off-flavor
(Aston and Dulley 1982). Alternatively, lactate can undergo anaerobic fermentation to
butyrate, H2 and CO2 (P. F. Fox and McSweeney 2004)
On the other hand, citrate is an important precursor of flavor compounds, and even though the
90% of the citrate in milk is lost in the whey, Cheddar cheese contains 0.2 to 0.5% that is
metabolized by LAB and NLSB to diacetyl, acetate, acetoin, 2,3 butandiol and CO2 (P. F. Fox,
Lucey, and Cogan 1990)
Lipolysis
Fat content has an important role in cheese flavor, and in addition to proportionate precursors
for the development of flavor compounds, and to be solvent for hydrophobic compounds, it
provides a fat-water-protein interface for catabolic reactions (P. Walstra et al. 1999). Indeed,
5
hydrolysis of lipids could be thought as the second most important metabolic reaction toward
the generation of flavor compounds in cheese.
Lipolytic activity is specific of the outer ester linkage of tri and/or diacylglycerides (Deeth
and Touch 2000a). Esterases and lipases hydrolyze acyl ester chains between 2 and 8 carbons,
and chains of 10 or more carbons respectively. The relative proportion of FFA in cheese from
C6:0 to C18:3 is similar to that in milk fat, but it is higher for free C4:0, evidence that it is
selectively release or synthesized by cheese microflora (Bills and Day 1964; P. F. Fox and
McSweeney 1998; Paul L.H. McSweeney and Sousa 2000). Lipolytic agents are indigenous,
endogenous and/or exogenous enzymes from: 1) milk; 2) rennet; 3) starter culture; 4)
nonstarter culture; and 5) adjunct cultures.
Milk contains a potent lipoprotein lipase (LPL) that in presence of an apolipoprotein activator
(C-II apolipoprotein-glutamic acid), releases sufficient free fatty acids (FFA) to gives a rancid
flavor to milk. However, LPL rarely reaches full activity due to its compartmentalization
along with fat and casein micelles, being surrounded by a lipoprotein membrane, which once
is damaged it may later promote off-flavors in cheese. LPL is relatively not specific for fatty
acids, but is for the sn-1 and sn-3 sites of mono, di and triacylglycerides, being selective for
short and medium chain triglycerides (Olivecrona et al. 1992). Due to LPL sensitivity to
temperature, its contribution to flavor generation is more important in raw-milk cheeses.
Indeed, the 73-95% of LPL activity is inactivated after pasteurization of milk (Yvonne F.
Collins, McSweeney, and Wilkinson 2003).
Usually commercial Rennets are free from lipase activity. However, Greek and Italian
varieties prepared with rennet paste, contains a potent lipase, pregastric esterase, which is
highly specific for short chain acids at the sn-3 position (P. F. Fox et al. 2000). Although
lipolysis occurs in most cheeses, it is more noticeable in varieties made using rennet paste
(longer
ripening
and/or
developing
of
secondary
flora).
In
Cheddar
cheese
estereolityc/lipolytic enzymes of LAB are the main lipolytic agents during cheese ripening.
These enzymes are intracellular and can hydrolyze esters of fatty acids, tri, di and
monoglycerides (Holland and Coolbear 1996; Chich, Marchesseau, and Gripon 1997; P. F.
Fox et al. 1999; Liu, Holland, and Crow 2001). Although lactococcus and lactobacillus spp
6
are weakly lipolytic, due to their high concentration during long ripening period, they become
significant in the final levels of FFA. Moreover, it has been reported that esterase activity is
higher than lipase activity for many lactobacilli and lactococcus strains, such as Lb.
helveticus, Lb. delbrueckii subps bulgaricus, Lb. delbrueckii subps lactis, Lb. acidophilus, Lc.
lactis subps lactis, and Lc. lactis subps cremoris (Tsakalidou and Kalantzopoulos 1992). Due
to LAB enzymes are intracellular, they are release by cell autolysis. And it has been reported
that in Cheddar cheese the use of autolytic strains such as Lc. lactis subps cremoris AM2,
result in higher levels of FFA as well as higher levels of secondary proteolysis than those for
the strain Lc. lactis subps cremoris HP (Wilkinson et al. 1994; Yvonne F. Collins,
McSweeney, and Wilkinson 2003).
FFA are precursors molecules for many catabolic reactions resulting in flavor and aroma
compounds such as methyl ketones, lactones, esters, alcanes and secondary alcohols (Gripon
et al. 1991; P. F. Fox et al. 1999; Paul L.H. McSweeney and Sousa 2000). Methyl ketones are
important flavor compounds generated from β-oxidations of fatty acids followed by
decarboxilation supported by a low redox potential and micro-aerophilic conditions. They are
very important and key compounds for the blue cheese flavor, yet they are identified as
impact compounds in other cheeses (Sablé and Cottenceau 1999; Qian and Reineccius 2002).
However, in full fat Cheddar cheese it has been reported that levels of heptan-2-one, non-2one, and undecan-2-one increase for 14 weeks and then decrease. In low fat milk Cheddar
cheese, methyl ketones levels correspond to 25% of those in full-fat cheeses (Dimos 1992;
Gerda Urbach 1993).
Esters are other product of FFA catabolism. They are highly flavored and arise from the
reaction between short to medium fatty acids and alcohols derived from the fermentation of
lactose. In Cheddar cheese some LAB cultures hydrolyze milk fat and esterifies certain short
fatty acids with ethanol, resulting in esters with fruity flavor notes such as ethyl butanoate,
(Molimard and Spinnler 1996a), which is considered as defect in Cheddar cheese (Paul L.H.
McSweeney and Sousa 2000). It has been reported that in Cheddar cheese esters of FFA are
ethyl derivates (Arora, Cormier, and Lee 1995).
7
Secondary alcohols are product from lipolysis, and result specifically from enzymatic
reduction of methyl ketones (Engels et al. 1997). In 1993 Urbach reported that in Cheddar
cheese, 2-propanol is product from acetone and 2-butanol from butanone. The odors can be
described as fruity, green, fuel oil like, and earthy, but the real contribution to the overall
cheese flavor is limited because they have high odor thresholds
Lactones are cyclic esters formed by esterification of hydroxy fatty acids. They confer a
nutty, coconut and buttery-type character to cheese (Wallace J.M. and Fox P.F. 1997; Dirinck
and De Winne 1999; Paul L.H. McSweeney and Sousa 2000). In addition, -lactones and lactones are present in milk and consequently in all types of cheese. Their concentrations are
correlated to the extension of lipolysis, and in Cheddar cheese it has been reported that levels
of lactones reach concentrations above their threshold during early ripening (Yvonne F.
Collins, McSweeney, and Wilkinson 2003b). However, high levels of lactones with high
molecular mass have been associated with rancid Cheddar cheese (Wong, Ellis, and LaCroix;
Jolly and Kosikowski 1975). -Lactones have higher detection threshold than -lactones, and
contribute to fruity notes such as coconut, peach and apricot (Dutossé et al., 1994; O´Keefe et
al., 1969).
It is proposed that cheese flavor is the balance of the contribution of many compounds present
at certain level. Even though, some varieties have specific major contributors, like methyl
ketones in mould ripening cheeses or FFA in hard Italian cheeses. However, for Cheddar
cheese, the exact role of individual compounds and the right balance among them is still
subject of research. Furthermore, in order to relate flavor of Cheddar cheese to levels of FFA,
research has been based on: 1) determination of individual levels of FFA; 2) addition of plain
bases to determine if they can be produced or improved; 3) selective removal of FFA to detect
any alteration in perceptible flavor; 4) manufacture of reduce fat cheese or cheese with
vegetable fat as substitute to milk fat (Wijesundera and Drury; Aston and Dulley 1982; Qian
and Reineccius 2002)
Proteolysis
Casein metabolism is the consequence of bacteria using proteins as substrate once lactose is
exhausted. This process contributes directly and indirectly to the development of flavor and
off-flavor compounds in Cheddar cheese. After protein breakdown into peptides and free
8
amino acids, catabolic reactions such as transamination, deamination, decarboxylation,
desulphuration and catabolism of aromatic amino acids, result in aroma compounds such as
volatile sulfur compounds, aldehydes, ketones, esters, and thioesters.
Proteolysis is the biochemical process that dominates the later phase of ripening (J. E.
Christensen et al. 1999; Tammam et al. 2000). The decomposition of the casein network
occurs due to the action of enzymes from the coagulant, the milk, the starter bacteria, nonstarter bacteria, and secondary cultures.
Traditionally, cheese is manufactured by using an enzymatic coagulant extracted from
abomasa of milk-fed calves known as rennet, and only up to 15% of the coagulant activity
remains after whey drainage (Upadhyay et al. 2004). Chymosin (EC 3.4.23.4) and bovine
pepsin (EC 3.4.23.1) are the major proteolytic enzymes in the coagulant and have a clotting
activity of 88-94%, and 6-12% respectively. Specifically Chymosin coagulates milk by the
rupture of the bond Phe105-Met106 in k-casein (Mulvihill and Fox 1979; P. F. Fox et al.
2000). In addition, Chymosin acts mostly on αs1-casein, by hydrolyzing the bonds Phe23Phe24 to form the peptides αs1-CN f24-199, αs1-CN f 1-23 and αs1 CN f102-109 (Richardson et
al. 1974; T.K. Singh et al. 1994). Hydrolysis of αs1-casein makes the texture of curd smoother
and homogeneous, and it has been acknowledged that increasing the salt in moisture level
does inhibits the subsequent hydrolysis of the peptide αs1-CN f24-199 (Exterkate, Alting, and
Slangen 1995). However, Chymosin activity on β-casein is extremely inhibited due to the
presence of NaCl, though the presence of the peptides β-CN-f1-192 and β-CN-f193-209 is
evidence of some enzymatic activity (S. ;Hup Visser 1983; Paul L. H. McSweeney et al.
1994). No major activity has been reported in αs2-casein and/or para-κ-casein (T.K. Singh et
al. 1994). Therefore with high ionic strength and low water activity breakdown of α s1-casein
is faster than that of β-casein (S. Visser 1993).
Among the most representative indigenous milk proteinase are Plasmin (the principal and
most studied one), Cathepsin D, Cathepsin B, and other proteolytic enzymes in lysosomes of
somatic cells such as the serine proteinase, Elastase.
9
Plasmin acts mainly on β-casein, and due to the complex system of activators and inhibitors of
the precursor Plasminogen, its enzymatic activity differs between cheese varieties. Plasmin
hydrolyses β-casein at the bonds Lys28-lys29, Lys 105-His106 and Lys107-Glu108, resulting
in the formation of γ1-[β-CN f29-209], γ2-[β-CN f106-209], γ3-[β-CN f108-209] caseins,
along with proteose peptones. The γ-caseins accumulate during ripening while proteose
peptones are hydrolyzed by starter bacteria peptidases which yield small peptides and free
amino acids (Tanoj K. Singh, Fox, and Healy 1995).
Cathepsin D, is an acid protease, with an optimum pH of 4.0, that produces the
glycomacropeptide κ-CN(f106-169) (P. F. Fox and McSweeney 1996). However, in spite of
Cathepsin D specificity is similar to that of Chymosin, Cathepsin D is not significant in milk
coagulation due to its milk clotting potential is fairly low (P. F. Fox and McSweeney 1996;
Larsen et al. 1996). Cathepsin D cleavage sites on αs1-casein and β-casein are similar to those
of Chymosin, but significantly different from those on αs2-casein (Larsen et al. 1996). In
addition, due to Cathepsin D is a heat labile enzyme, it is important in proteolysis of dairy
products made with pasteurized milk and in proteolysis of rennet free cheeses. Reason why
Cathepsin D activity during proteolysis of cheddar cheese is difficult to quantify (Hurley et al.
1999; V. Crow, Curry, and Hayes 2001)
Cathepsin B is a lysosomal cysteine proteinase that works in the degradation of proteins by the
cell. It is activated by dithiothreitol (DTT), has a pH optimum of 6.0, partially survives to the
pasteurization processes, and is capable of degrading αs1-casein and β-casein extensively
(Knecht 1999; Considine et al. 2004). Cathepsin B proteolytic activity is of great interest due
to the activity of lysosomal enzymes has been related to the poor quality of dairy products
(Grandison and Ford 1986; Verdi and Barbano 1991)
Elastase is a neutral serine proteinase, and its essential physiological function is the
degradation of elastin. However, it has a broad specificity on αs1-casein and β-casein, with a
preferred specificity for bonds involving uncharged, non-aromatic amino acids (Naughton and
Sanger 1961), cleaving 25 and 19 sited respectively (Considine et al. 2000). On β-casein,
some of the Elastase cleavage sites are identical to, or near to those cleaved by Plasmin,
10
Chymosin or cell envelope-associated proteinase of several strains of Lactococcus (Considine
et al. 1999).
Regarding to the starter lactic acid bacteria (LAB), the principal starter cultures used in cheese
manufacturing are the mesophilic Lactococcus and Leuconostoc species, and the thermophilic
Lactobacillus and streptococcus thermophiles species. Their main role is to decrease the pH
by producing lactic acid from lactose. Lactic acid bacteria (LAB) possess a complex
proteinase/peptidase system, very important during the secondary proteolysis, which includes:
1) a cell envelope proteinase, lactocepine; 2) the intracellular oligoendopeptidases PepO and
PepF; 3) the general aminopeptidases PepN, PepC, PepG, along with the glutamyl
aminopeptidase (PepA), the pyrolidone carboxylyl peptidase (PCP),
the leucyl
aminopeptidase (PepL), the X-prolyldipeptidyl aminopeptidase (PepX), the proline
inmunopeptidase, the aminopeptidase P (PepP), the prolinase (PepR), and the prolidase
(PepQ); 4) the general dipeptidase PepV, PepD, PepDA and the general tripeptidase (PepT)
(Sousa, Ardö, and McSweeney 2001). These peptidases can be classified into: 1)
endopeptidases, relevant for the degradation of oligopeptides to shorter peptides; and 2)
exopeptidases such as carboxypeptidases or aminopeptidases, which release free amino acids
from short peptides. The most important enzyme is lactocepine, a serine proteinase that
degrades intermediate size peptides produced from Chymosin and Plasmin activity (Upadhyay
et al. 2004). However, since caseins are rich in proline and because of its particular structure,
proline specific enzymes have significant contribution to the proteolysis of cheese, by making
released peptides accessible to other peptidases.
Although the role of non-starter lactic acid bacteria (NSLAB) in the development of the flavor
of cheddar cheese is not totally understood, it has been used to manipulate the final sensory
characteristics of the product. Indeed, when LAB population declines, NSLAB population
becomes the dominant microflora in the maturation of cheese (Peterson and Marshall 1990; V.
L. Crow et al. 1995; P. Fox, McSweeney, and Lynch 1998). NSLB are essentially constituted
by homo and hetero-fermentative species of lactobacilli and are determinant during the
secondary proteolysis. However, hetero-fermentative and certain Lactobacillus strains have
been associated with off-flavors in cheddar cheese (Puchades, Lemieux, and Simard 1989),
and it has been reported that thermophilic lactobacilli do not influence the development of
11
cheddar flavor (Lloyd, Horwood, and Barlow 1980). Actually, it has been suggested that the
addition of selected adjunct strains of Lactobacillus spp, positively influence the quality of
cheese (Drake et al. 1996; C.N. Lane and Fox 1996a; C. M. Lynch et al. 1996; Muir, Banks,
and Hunter 1996), and accelerate the ripening of standard and reduced-fat Cheddar cheese (M.
A. El Soda 1993; Christensen J.E., Johnson M.E., and Steele J.L. 1995) Thus, due to there are
not criteria for the selection of adjuncts, research has been done to clarify the proteolytic and
lipolytic systems of NSLAB which contribute to the maturation
FACTORS AFFECTING THE DEVELOPMENT OF VOLATILE
COMPOUNDS
Cheese flavor development and most of its chemical and physical properties relates to
microbiological and biochemical events that depend on physicochemical parameters,
established during the early stages of cheese manufacture. Factors such as pH, water activity
(aw), buffer capacity, oxidation-reduction potential (redox), physical history of milk and size
and geometry of curds are set during the manufacturing of curd, which as mentioned above it
is a dehydration process in which fat and caseins are concentrated 6 to 12 fold through the
following operations:
1) Preparation of milk (pasteurization);
2) Acidification (LAB);
3) Rennet coagulation;
4) Syneresis;
5) Pressing and shaping the curd;
6) Salting;
7) Other operations.
However, other physical factor such as time and temperature are relevant dynamic factors
during the ripening stage that are of great interest in the discussion of cheese flavor.
12
Milk preparation: selection, pasteurization & standardization
After refrigeration of raw milk, microflora is dominated by psychrothrops which produces
heat stable lipases and proteinases, and in concentrations over 106 CFU/ml, they reduce yield
or can cause development of off-flavors during ripening (Payne and Kroll 1991)
Mainly, because of heat-induce changes in cheese micro-flora, cheese made from raw milk
ripens faster and develops stronger flavors than those made from pasteurized milk (P. F. Fox
and McSweeney 1998). In addition to the changes in microflora, heat treatments of milk, such
as pasteurization or heat shock, causes inactivation of indigenous enzymes (mainly lipoprotein
lipase inactivation), resulting in cheeses with weaker flavor and slower ripening (P. F. Fox and
McSweeney 1998; Y.F. Collins, McSweeney, and Wilkinson 2004).
On the other hand, standardization is a process through which is possible to control the ratio of
casein to fat, influencing in this way parameters such as moisture and moisture in non fat
substances, improving cheese yield.
Some of the different methods to pre-determine the composition of milk to assure a
homogenous production during the year involve the use of membrane concentration
techniques or vacuum concentration techniques. In the case of cheddar cheese (full fat and
reduce fat cheeses), most of the research work is related to vacuum condensation, from where
industrial and academic results show that concentration cannot exceed 1.8:1, because at higher
concentration sweetness and saltiness problems occur due to excessive lactose and minerals
(Anderson et al. 1993). Contrary, membrane concentration techniques (reverse osmosis,
nanofiltration, microfiltration and ultrafiltration) are common in cheddar cheese production.
Indeed, low concentration or protein standardization by ultrafiltration is a popular method to
secure the uniformity of milk composition, lower casein loss through a firmer curd, more
efficiency and better yield (Mistry and Maubois 2004). Additionally, different dairy products
such as milk powder, milk protein, milk permeate can be used as means to standardize
components in cheese milk (Mistry, Metzger, and Maubois 1996).
13
Acidification
In addition to initiate the fermentation of lactose to lactate, during the acidification, starter
cultures provide enzymes for the ripening stage, promoting as well the prevention of spoilage
because of the reduction of redox potential of the system.
Due to Lactic acid bacteria (LAB’s) are auxothropic for many amino acids, they have a
proteinase and peptidase systems to liberate amino acids from caseins. Also, they poses
intracellular metabolic enzymes which catalyze the catabolism of amino acids and contribute
to the formation of volatile flavor compounds (Yvon and Rijnen 2001; Y.F. Collins,
McSweeney, and Wilkinson 2004; P. F. Fox and McSweeney 2004).
Consequently the reduction of pH due to the fermentation of lactose to lactate and lactic acid
determines:
1) The retention of coagulant activity in the curd;
2) The rate of Syneresis;
3) The growth of added and native microorganism;
4) The enzymatic activity during ripening.
LAB for cheddar cheese is a mesophilic culture with optimum temperature for growth at 2030 C, so when the cooking temperature is higher than the optimum the rate of syneresis is
increased but the acid development decreases.
Coagulation
As it was mentioned above, it occurs via limited proteolysis of k-caseins at or near Phe105Met106 bond followed by the aggregation of the Ca2+ rennet altered micelles at temperature
over 18 C (Home and Banks 2004).
The traditional rennet is a brine of the abomasa of milk fed calves (or other young dairy
animal), which contains mainly Chymosin and low levels of pepsin. The retention of the
rennet activity added to milk varies from 0-15% depending on factors such as type of enzyme,
pH at whey drainage, cook temperature, and moisture content of the curd (Upadhyay et al.
14
2004). Thus, later on during the ripening stage, parameters for coagulation will influence the
degree of hydrolysis of αS1-casein, the generation of flavor compounds, texture of curds and
cheeses, and subsequently the liberation of sapid compounds.
Ph determines the casein micelle matrix, and along with other factors such as heating, protein
concentration in milk, and addition of Ca2+, it establishes the equilibrium between colloidal
and dissolved calcium phosphate and between dissolved calcium phosphate and other ions.
For example acidity reduces the negative charge on the micelles (caseins isoelectric point is
4.6) and increases solubility of citrates (pH 5.5 completely soluble) and colloidal calcium
phosphate (pH 5 completely soluble). Heating reduces dissolved calcium and phosphates, and
promote coagulation by association of colloidal phosphates with the casein micelles. Milk
concentration increases buffer capacity and colloidal calcium phosphate and its association to
the micelles.
Therefore, it is possible to observe how coagulation depends on dynamic properties like pH
and the solubility of calcium salts, and the cleavage of κ-caseins by rennet enzymes. Thus,
defects such as sour and/or bitter flavor and soft and pasty body are associated with excessive
acidity and a pH under 5.0.
Syneresis
The gel formed due to the rennet induced coagulation is stable, but once it is broken, most of
the liquid entrapped in the gel is expelled readily as whey. Syneresis and consequently
moisture content of cheese are controlled by the milk composition, size of the curd particles,
cooking temperature, time, rate of acidification and rate of stirring the curd-whey mixture (P.
F. Fox and McSweeney 2004). Therefore, the mentioned parameters and the operations
involved with Syneresis have an impact on ripening. Reason why high moisture cheese ripens
faster than low moisture cheese. In addition to promote syneresis, high cooking temperature
inactivates the remaining Chymosin and increases the levels of Plasmin due to denaturation of
Plasmin inhibitors and inhibitors of Plasminogen activators (Farkye and Fox 1990).
High temperature during rennet gelation increases the initial porosity of the gel and
consequently the initial rate of syneresis after the curd is cut. The rate of syneresis is also
15
increased by smaller curd size at cutting and a vigorous agitation (Pieter Walstra, Wouters,
and Geurts 2006).
Cheddaring
During cheddaring the pH of curds decrease to 5.4, causing dissolution of colloidal calcium
phosphate, which modifies the texture of the curd by altering Ca:protein ratio. Some of the
objectives of the cheddaring process are to remove small amount of whey to allow a better
acid development, to have moisture control, to develop a proper texture, and curiously to
repress the growth of gas forming spoilage organisms.
Salting
Salting can be achieved by immersion in brine (most chesses), by mixing dry salt with the
milled curd, which is the case of Cheddar cheese, or by application of salt on the surface after
molding. The brine option is intended for cheeses with high level of salt and the rate at which
salt in moisture increases is slow since NaCl should diffuse from the surface. Contrary in the
dry salted varieties NaCl uptake is very fast (Guinee 2004). This process influence the flavor
profiles of cheese by controlling microbial growth, determine enzyme activity and impact on
water activity.
Salting affects bacterial and enzymatic activity. Indeed, inhibition of acid production occurs at
NaCl concentrations over 1.5%, which has been used in some cheese varieties to stop
acidification and to fix the pH (Guinee 2004). In addition, it is well known that microbial
enzymes can be inhibited or stimulated by moderately high NaCl levels, particularly at low
pH, depending on the type of enzyme. As matter of fact, approximately 2 kg of H2O are lost
by absorption of 1 Kg of NaCl , which means a reduction on the water activity of the system.
For example, Plasmin (principal indigenous proteinase in milk) is stimulated by 2% NaCl but
inhibited at high concentration (Farkye and Fox 1990), but for Chymosin (principal enzyme in
the rennet), the addition of NaCl increases the ionic strength, promoting interactions between
hydrophobic C-terminals of β-caseins, inhibiting the access of the enzyme to cleavage sites.
Thus decreasing NaCl levels facilitates Chymosin action on β-caseins and production of
hydrophobic peptides.
16
In the case of cheddar cheese most of the acid is developed before salting and molding,
therefore cooking time and temperature along with culture activity determine curd pH, buffer
capacity and rennet retention at draining.
Pressing and molding
Curds for brine salted are molded prior to salting while those for dry-salted are molded and
pressed after addition of salt.
Cheese size is not determined only due to esthetic reasons, contrary because of many
biochemical events occur at the surface of the cheese during the ripening stage, thus
controlling the ratio of surface area to volume is key to control and assure an homogenous
ripen, and depending on the variety, it allows the development of certain characteristics such
as eyes formation due to sufficient partial pressure of CO2. In the case of cheddar the time
and temperature of pressing and the temperature during the first days of ripening influence the
extent of acid development and the minimum pH.
Other important factor involved with flavor development is the oxidation/reduction system of
milk,
which
include
Fe2+/Fe3+,
Cu+/Cu2+,
dehydro
ascorbate,
rivoflavin,
and
lactate/pyruvate (P. Walstra et al. 1999). By decreasing the pH and increasing temperature free
sulfydryl groups are produces in proteins and amino acids lowering the redox potential.
Usually the redox potential decreases during setting, rises during, cutting, cooking and
draining, decreases during cheddaring, increases during milling, and decreases during pressing
and ripening. Maintaining a low Eh (-150 to -300 mV) promotes the production and stability
of sulfur volatile compounds (G. Urbach 1995; Beresford et al. 2001)
Therefore, after the early stages of cheese making, the main factors that determine the
structure and flavor of most varieties are: 1) the extent of acid production in the vat; 2) amount
of starter culture 3) the residual plasmin and rennet: 4) the pH of curds at draining; 5) the
residual lactose; 6) and the mineral content of the curd.
17
CHEESE FLAVOR AND FLAVOR ANALYSIS
As a response to the need for exact and predictable results of the ripening stage, monitoring
the ripening process has been of interest for cheese makers and food scientist, especially when
large productions are involved and “standardization of flavor” is associated to high quality
products by the consumer. Thus, different methods to monitor the extension of ripening have
been developed, and this work will be based on a metabolic approach.
From this perspective, ripening can be followed by determining the metabolite products from
primary carbon metabolism paths such as 1) the conversion of lactose and citrate; 2) lipolysis;
3) and proteolysis. Since the degradation and conversion of caseins is the most important
biochemical step for flavor formation in hard and semi-hard cheeses, tracking the progress of
proteolysis is an excellent indicator of the ripening process.
There are different analytical techniques to monitor patterns of proteolysis; one of them is a
fractionation scheme to extract nitrogen compounds based on pH. Nonetheless, choosing the
scheme and methodology depends on 1) the availability of equipment and resources; 2) the
cheese variety and its characteristics; and 3) the objective of the study. (Ardö and
Frederiksberg 1999; Sousa, Ardö, and McSweeney 2001). These fractions can be used to
obtain information about one or more proteolysis agents, such as action of certain enzyme,
comparison of coagulants, and analysis of different starters or adjunct culture. The insoluble
fraction can be used to study effects on primary proteolysis by means of Urea-polyacrylamide
gel electrophoresis or by capillary electrophoresis. As well insoluble and soluble fractions can
be used to determine peptide profiles by reverse phase high performance liquid
chromatography (RP-HPLC), and to perform an analysis of individual amino acids. However,
it is important to remember that proteolysis is too complex to be described by a single index,
and in addition other chromatographic techniques involving mass spectrometry such as HPLC
MS-MS or GC-MS are a more powerful approach for identifying and monitoring specific key
flavor and taste compounds in a fast and accurate way (Cserháti 2002; Careri and Mangia
2003).
Regarding the cheddar cheese making process, other metabolites could be tracked too;
however, not all of them can be use because of certain limitations related to the metabolic
18
path. For example, the remaining lactose in cheddar cheese cannot be considered as a relevant
parameter to monitor the ripening progress due to the remaining lactose is metabolized to Llactic acid, ethanol and CO2 within the first few weeks. And in spite that ethanol can esterified
with free fatty acids to produce ethyl-esters, compounds associated to fruity notes, they are not
good descriptors either. On the other hand, despite lipolysis is very limited for this variety, it
is an important metabolic path to track, where the low but steady increased in FFA’s,
especially those levels for C4 to C8 ones, can be use as an indicator of the progress of the
ripening (Yvonne F. Collins, McSweeney, and Wilkinson 2003b; Y.F. Collins, McSweeney,
and Wilkinson 2004). However, it is important to keep in mind that this last one as well as
other indicators of proteolysis does not correlate directly to the production of relevant volatile
compounds.
Furthermore, the aroma analysis highly relies on the technique for extraction and
concentration use to isolate the different aromas compounds. And due to there is not a single
universal method to extract all kind of aroma compounds at once, many techniques has been
used to study the aroma compounds of cheddar cheese, such as: solvent extraction,
simultaneous distillation extraction, static and dynamic head space, purge and trap, ion
exchange chromatography and solid phase microextraction. Additionally, for works intended
for tracking the progress of ripening, the methodology selected should focus on the expected
results as consequence of the assessed agent or parameter.
By using the methods mentioned above, a wide variety of compounds such as acids, esters,
ketones, aldehydes, alcohols, lactones, phenols and volatile sulfur compounds have been
identified in cheddar cheese. Most volatile aroma compounds are found in the lipid content
because they are hydrophobic. Thus, in the case of fatty acids it is relevant to know the
individual concentration because each one has different sensory threshold and aroma attributes
and contribution. They can be determined by gas chromatography (GC) without derivatization,
but they require separation from triglycerides and others lipids such as cholesterol and
phospholipids, prior to GC analysis. This is done by the use of aminopropyl weak anion
exchange columns, in which the 100% of the recoveries have been achieved and a good
repeatability can be expected (De Jong and Badings 1990; Chavarri et al. 1997; Qian and
Reineccius 2002).
19
Based on the research done on amino acids catabolism (D. J. Manning, Chapman, and
Hosking 1976; B. Weimer, Seefeldt, and Dias 1999; Stuart, Chou, and Weimer 1999),
particularly by LAB enzymes, breakdown of key compounds such as phenylalanine
(aromatic), leucine (branched chain) and sulphurous (methionine and cysteine), have been
characterized. Thus, volatile sulphur compounds (VSC’s) such as methanethiol, dimethyl
sulphide, dimethyl disulphide and dimethyl trisulphide, most of them with low odor threshold,
are important flavor compounds that can be used as tracking metabolite for monitoring the
progress of the ripening stage.
VSC’s cannot been studied by conventional techniques including static head space and purge
and trap due to the loss of analyte during concentration stage and the potential formation of
thermal artifacts, particularly in these case where the studied compounds are highly volatile
and chemically reactive. A better approach is the solid phase microextraction technique
(SPME) along with the a sulphur pulsed photometric detector (PFPD), which makes possible
the analysis of VSC’s in food faster and without tedious sample preparation involving the use
of solvents (Fang and Qian 2005; Vazquez-Landaverde, Torres, and Qian 2006; H. M.
Burbank and Qian 2005) . Nonetheless, there are many kind of fibers, but the most appropriate
type for the study of dairy products is Carboxen-polydemethylsiloxane (CAR-PDMS), which
can readily extract high volatile and low molecular compounds including VSC’s
Instrumental analysis of cheddar cheese
Chromatography methods are essential in understanding the aroma chemistry of cheddar
cheese, and consequently they offer a thorough picture of the impact of manufacturing
variables during the production process on the ready to sell product. Some of the preferred
methods are gas chromatography (GC), including the use of detectors such as flame ionization
detector (FID), pulsed flame photometric detector (PFPD) and mass spectrometry (MS)
among others, and high performance liquid chromatography (HPLC), operating in the reverse
phase separation mode. However, cheese matrix is complex and samples cannot be injected
directly into the injection port of the devices, and as consequence isolation of volatile, nonvolatile and/or sapid compounds is required in order to protect the instruments and to avoid
degradation of samples and formation of artifacts due to the high temperatures in the GC
20
injection ports. Thus, to achieve meaningful and reproducible results, isolation and
concentration of aroma compounds is mandatory, and depending on the chemical composition
of the target analytes, different approaches can be use to perform their separation from the
matrix. Therefore, procedures involving pH, solubility, polarity and volatility are often used in
the aroma analysis of cheddar cheese.
Gas Chromatography
The separation, identification, quantification and analysis of volatiles is generally complicated
due to their concentration levels, which could be as low as parts per billion (ppb) in the
nonvolatile matrix. The most common isolation methods used for volatiles are solvent
extraction and distillation, and headspace techniques, including static headspace analysis,
dynamic headspace analysis and headspace-solid phase microextraction (SPME).
Extraction-Distillation
Cheese is usually low moisture, high fat and high protein product, and in spite of that difficult
matrix, solvent extraction became a common procedure to separate volatiles. It involves the
use of organic solvents such as acetonitrile, dichloromethane and diethyl ether, where the last
one is considered as the most suitable for cheese analysis due to its low density, low boiling
point and high selectivity for aroma compounds. Moreover, purity of solvents is critical,
which implies an obligatory pre-distillation/purification process before its use. In the case of
cheddar cheese and other hard and semi-hard cheeses, the samples are usually frozen
employing liquid nitrogen, then grated or grinded, and finally the analytes are extracted with
the selected solvent (Preininger and Grosch 1994; Milo and Reineccius 1997; Suriyaphan et
al. 2001; Zehentbauer and Reineccius 2002; Qian and Reineccius 2002; Avsar et al. 2004;
Mary E. Carunchia Whetstine, Cadwallader, and Drake 2005), or alternatively, an aqueous
extract is prepared followed by solvent extraction (Moio et al. 1993). However, because
cheese is not a free-fat product, an additional extraction is required and usually done by
dialysis (Benkler and Reineccius 1980) or by a low-temperature high vacuum distillation
(Suriyaphan et al., 2001) to separate non-polar and non-volatile lipids, and other non-volatile
constituents, followed by the separation of analytes into neutral, basic and acidic fractions if it
is necessary.
21
The approach through direct solvent extraction is effective to separate semi-volatiles such as
lactones, free fatty acids and phenolics, but has some disadvantages such as evaporation of
large quantities of solvent, being a slow process, and adsorption of components of the
membrane surface and catalysis of acetone condensation in the case of the dialysis (Benkler
and Reineccius 1980).
Regarding distillation as method to purify the solvent extracts, high vacuum distillation and
steam distillation techniques are suitable when the sample will be analyzed by GC on capillary
column. The most popular techniques are simultaneous steam distillation extraction (SDE)
(Chaintreau 2001) and Solvent assisted flavor evaporation distillation (SAFE) (Engel, Bahr,
and Schieberle 1999). The first one has been used for the analysis of aqueous slurry of
Cheddar, Gouda, Edam, Swiss and Parmesan cheese, with the Likens-Nickerson extractor
using diethyl ether, and following a procedure described by Parliament (1998). However, this
technique can lead to artifact formation due to it is performed at high temperatures. Regarding
SAFE, it has been reported that the method resulted in better recovery than the traditional high
vacuum transfer, and it was effective for most volatiles from solvent extracts of cheese
(Werkhoff et al., 2002; Carunchia-Whetsine et al., 2005, 2006; Cadwallader et al., 2006;
Schlichtherle-Cerny et al., 2006)
Headspace methods
As it was mentioned above, headspace methods include static headspace analysis (SHA),
dynamic head space analysis (DHA) and headspace-solid phase microextraction (SPME),
which are limited to the equilibrium of volatile compounds into the gas phase, reason why
they are considered non destructive and require minimal sample preparation.
SHA is used when analysis of major component is satisfactory. In this technique, the sample is
contained in a closed vessel and volatiles are allowed to reach the equilibrium by partition into
the head space and the matrix, which is a process influenced by temperature, vessel size, ratio
of sample to head space volumes, addition of salt, and agitation. Then an aliquot of the
headspace is taken and injected into the GC. SHA has been used for the analysis of highly
volatile sulfur compounds such as hydrogen sulphide, methanethiol and dimethyl sulphide
(Lin and Jeon 1985), acetaldehyde and other low molecular weight Strecker aldehydes such as
22
methyl propanal, 2-methyl butanal, and 3-methyl butanal (Fernández-García 1996), which are
important contributors to the cheddar cheese flavor. However, this technique lacks of
sensitivity and is really difficult to standardize.
By using a flow of carrier gas and an intermediate adsorption or cryogenic step before the
injection into the GC, larger amounts of volatiles can be collected and the efficiency of the
headspace analysis is improved, turning it into Dynamic headspace analysis (DHA) or purgeand-trap analysis. Usually the carrier gas could be helium or nitrogen, while the adsorbent
might be a polymeric material such as Tenax (poly-2, 6-diphenyl-p-phenylene oxide) or active
charcoal. Adsorbent trapping is more common than cryogenic focusing, because it prevents
the damage of the GC column. Once the volatiles are isolated, they are desorbed by heating,
they could be concentrated by cryo-focusing, and then they are transferred to the GC column
by thermal desorption. Some of the advantage is the low risk of artifact formation, the minimal
sample preparation and the fact that it is solvent-free. However the principal limitation is that
the method is not efficient for semi-volatiles. Extracts by this technique have been analyzed by
multidimensional CG-O/FID/MS and in total 5 aldehydes, 6 ketones, 8 alcohols, 3 esters, 11
hydrocarbons, 3 halides and 3 sulfur compounds were identified (Arora et al., 1995/ paper;
Dunn and Lindsay, 1985; Barbieri et al, 1994; Thierry et al, 1999, 2004; Larrayoz et al, 2001;
Rychlik and Bosset, 2001a; Valero et al., 2001; Qian and Reineccius, 2002; Boscaini et al.,
2003; Avasar et al., 2004).
Another solvent-free isolation method is SPME, which can concentrate volatiles from various
matrices in a single step (Kataoka and others, 2000). It reduced the time required for sampling
and the cost of analysis, and it became the most common method for analysis of volatiles in
the last decade. It is based on the partition of the analytes in the headspace and the polymer
coated fiber. The method counts with great selectivity and specificity due to the several
adsorbents phases and film thicknesses available. The adsorption depends on temperature,
vessel size, ratio of sample to head space volumes, addition of salt, agitation, nature of the
coating and exposure time, for which is recommended short times (1-5 min) for highly
volatiles and (5-30 min) for semi-volatiles (Roberts et al., 2000). Volatiles are transfer to the
GC by thermal desorption in the splitless injection mode. Cheddar cheese was initially
analyzed using fibers coated with non-polar polydimethylsiloxane (PDMS) and polar
23
Polyacrylate (PA), showing better results for the polar coating (Chin et al., 1996). Volatile
components such as fatty acids and δ-lactones were found, and differences in profiles were
observed between varieties. Later, the technique was consolidated for the analysis of cheddar
cheese
aroma,
by
a
comparison
polydimethylsiloxane/divinylbenzen
of
five
types
(PDMS-DVB),
of
PA,
fiber
coatings,
carboxen/PDMS
PDMS,
and
carbowax/DVB. The study revealed that bipolar coatings PDMS/DVB and carboxen/PDMS
showed better selectivity, measured by the amount of peaks absorbed, besides confirming the
relevance of the adsorption temperature and exposition time (Dufour et al., 2001). Initially it
was suggested that unlike solvent extraction techniques or DHA, quantification of volatiles
was not possible, however it was demonstrated the fact that it was possible to perform a
quantification of volatiles extracted by SPME.
Separation
The injection technique and the analytical column stationary phase could be considered as the
most critical parameters influencing the result of the GC analysis. Regarding the first one,
programmable temperature vaporizer (PTV) and on-column injectors are the best options for
analysis of aroma extracts. They provide versatility, which make them even suitable for works
with cryogenic focusing from DHA, and ramped heated injections in the splitless or split
modes. Besides, they proportionate the possibility of avoiding thermal degradation in the cool
on-column injection for extracts from direct solvent extraction coupled with high vacuum
distillation. On the other hand, polarity of the stationary phase determines the type of analytes
that could be identified. Thus, DBWAX and FFAP phases constitute the adequate options for
analysis of polar compounds, while DB-5 phase is the right choice for non-polar compounds.
To study aroma compounds that make a contribution to the odor of a food, there are different
approaches including gas chromatography-olfactometry (GCO), odor activity value (OAV)
calculation, and sensory analysis. In GCO, the most often used techniques are the dilution
analysis, where the most popular ones are the aroma extract dilution analysis (AEDA)
(Grosch, 1993), its variation called aroma extract concentration analysis (AECA) (Kerscher
and Grosh, 1997) and GCO-headspace dilution analysis. In AEDA a dilution series, or
concentration series in the case of AECA, of an aroma extract is evaluated by GCO, where
compounds are ranked according to its potency based on the highest dilution or concentration
that can be perceived, which is defined as flavor dilution factor (FD). However, FD factors do
24
not account for highly volatile compounds lost during extraction and concentration,
underestimating important compounds. GCO-headspace dilution analysis provides a
complementary evaluation of the aroma composition. And in these methods, dilutions are
achieved by decreasing the headspace volume, or the purge gas volume in the case of the
DHA.
In addition, retention indices are used for confirming a proper identification of compounds
contributing to the aroma of a food. These are based on the retention times of target
compounds compared to those of certain standards (usually hydrocarbons). In addition, OAV
relates analytical data to sensory characteristics of a sample, and essentially it establishes a
relation between the concentration of a target compound in the food to its odor threshold;
allowing to discriminate accurately the compounds that really contribute to the specific aroma
of a food.
To the date, something to keep in mind is that the methodologies mentioned above coupled
with the separation efficiency of gas chromatography and the excellent identification potential
of mass spectrometry, offer to scientists and engineers the possibility to understand, describe
and improve production processes towards the satisfaction of customers, by maintaining
constancy in the quality of a product, in this case the traditional aroma of Cheddar cheese.
Reverse Phase-High Performance Liquid Chromatography (RP-HPLC)
The assessment of proteolysis through peptide profiling of cheese extracts by RP-HPLC, is
widely used to study quality and authenticity of samples. It is based on the principle that
caseins might not be soluble in certain solvents, but peptides resolved from their degradation
can be. What is evident when the amount of intermediate and small peptides increase as the
proteolysis of caseins takes place during ripening cheese (Bansal, Piraino, and McSweeney
2009; Piraino, Parente, and McSweeney 2004).
Most of the HPLC separations are done in the reverse-phase mode, recognized by the use of a
non-polar stationary phase and a polar mobile phase, where solutes are mainly retained due to
hydrophobic interaction with the non polar one, and are eluted in order of decreasing polarity.
Indeed, solutes retention decreases by increasing the organic solvent content of the mobile
25
phase. And as in the case of GC, samples cannot be directly injected in to the instrument,
meaning that isolation of analytes is previously required.
The most common way to perform the extraction of peptides from cheese, is following a
fractionation scheme that involve the use of solvents such as water, buffers at pH 4.6,
Trichloroacetic acid (TCA), phosphotungstic acid (PTA), and ethanol (Ardö and
Frederiksberg 1999). This, in addition to be a sort of sample preparation, allows measuring of
proteolytic activity through the analysis of nitrogen content of resolved fractions, and RPHPLC analysis of water and pH 4.6 soluble extracts. As a matter of fact, fractionation with
water is a method use for mature cheddar cheese due to its low and relatively constant pH
(Ardö and Frederiksberg 1999). Furthermore, water soluble nitrogen extracts (WSN) have
been used as index of ripening, which is only employed when there is no variation of pH
during ripening or between samples. Whereby, Bansal and others (2009), have suggested that
pH 4.6 soluble nitrogen should be used instead of WSN as index of primary proteolysis.
However, it is important to keep in mind that results from this method are somewhat smaller,
it is more difficult to perform, but is easier to standardize. On the other hand, because cheese
is a dynamic system, meaning that results of analysis depend on the age of samples (Sousa,
Ardö, and McSweeney 2001), RP-HPLC analysis of water and/or pH 4.6 soluble extracts can
be used as a discriminant technique for the characterization of secondary proteolysis.
Regarding TCA, it has been used to precipitate peptides from water and pH 4.6 soluble
extracts, at concentration ranging from 2 to 12%, which depends on the degree of fractionation
required (Bansal, Piraino, and McSweeney 2009). It has been reported that during the
fractionation with TCA, parameters such as extraction time and extraction temperature have
little or none effect on the amount of nitrogen obtained. In contrast, the amount and type of
peptides that can be extracted vary according with parameters such as cheese to water ratio,
pH, NaCl content of cheese and type of previous fraction (Polyachroniadou et al., 1999
chapter). Currently, ethanol, ranging from 30 to 80%, has been used as an alternative to TCA
because it offers similar precipitation levels and it can be easily evaporated for further analysis
of peptides in this fraction. Phosphotunstic acid (PTA) is a very discriminant protein
precipitant that is employed in a range from 1 to 6.5%, where only free amino acids (expect
lysine and argentine), and peptides under 600 Da are soluble. However, there are other
26
techniques to perform the fractionation based on peptides molecular mass rather than on
solubility such as dialysis, ultra filtration (UF), and size exclusion chromatography, which are
really useful for taste panel work due to they are solvent free (Fox, 1989 chapter).
Separation and characterization of peptides has been done by RP-HPLC analysis of WSN
extract, pH 4.6 soluble and insoluble extracts, 10 kDa UF permeate, 70% soluble and
insoluble extracts, and fractions from gel permeating chromatography. The elution of analytes
takes place in a Nucleosil RP-8 (250 × 4.6 mm, 5 μm particle size, 300 Å pore size) analytical
and guard columns (4.6 × 10 mm), and is usually done in the gradient mode using system such
as water/acetonitrile or water/methanol (Allan J. Cliffe, Marks, and Mulholland 1993);
however it is possible to perform it by isocratic elution using a phosphate buffer as mobil
phase (Pham and Nakai, 1984). The detection of analytes is done by monitoring the carbonyl
group in the peptides bonds, using a UV detector at wavelength ranging from 200 to 230 nm,
for which the most used ion-pair reagent is trifluroacetic acid (TFA).
Due to peptide profiles are multivariate in nature, the identification of analytes from raw data
has been done by visual matching (A.H Pripp et al. 1999; Are Hugo Pripp, Stepaniak, and
Sørhaug 2000), or by division of chromatograms in classes of retention time and peaks in each
class (Barile, 2006) followed by mass spectrometry. Whereas, data analysis has been done by
descriptive techniques such as Principal Component Analysis (PCA) (A.H Pripp et al. 1999;
Are Hugo Pripp, Stepaniak, and Sørhaug 2000; Piraino, Parente, and McSweeney 2004), or by
descriptive and inferential techniques such as Linear Discriminant Analysis on Principal
Component Scores (Hynes et al., 2003) and (O’Shea et al., 1996), Partial Least Squares
Regression (PLSR) and/or Partial Least Squares Discriminant Analysis (PLSDA).
Electrophoresis
Besides chromatographic methods such as GC and RP-HPLC, another analytical method for
assessment of proteolysis is electrophoresis, which is a specific technique that in addition to
give information about the extent of proteolysis and the general contribution of proteolytic
agents, also resolves, isolates and identifies peptides (Bansal, Piraino, and McSweeney 2009).
Indeed, it is a technique applied to the study of cheese ripening, and is particularly useful for
27
the comprehension of primary proteolysis, since it is limited to monitoring hydrolysis of
parent caseins, where only protein and large peptides can be visualized.
Among the different electrophoretic methods used in the study of cheese ripening, it is
possible to find works with paper electrophoresis, free boundary electrophoresis, high voltage
paper electrophoresis, starch gel electrophoresis, isoelectric focusing, capillary electrophoresis
(CE), and polyacrilamide gel electrophoresis (Fox 1989; Mcsweeney and Fox 1993; Fox et al,
1995; Bican and Spahni, 1993; Trieu-Cuot and Gripon, 1982; Addeo et al 1983, 1990; Moio,
Luccia and Addeo 1989, 1992; Creamer, 1992; Amigo et al, 1992; Strange et al 1992; Ledford
et al., 1966; Shalabi and Fox 1987; Andrews 1983; Blakesley and B, 1977; Lindeber 1996;
Goulds Worthy et al 1999; Bansal, Piraino, & McSweeney, 2009). The last one is the most
common method applied, it usually works with discontinuous buffer systems, utilizing urea or
sodium dodecyl sulfate (SDS) as dissociating agents. However, in the case of cheese analysis,
due to caseins have similar molecular weights, it has been reported that buffer systems using
urea are more sensitive and adequate than those using SDS, facilitating in this way to resolve
proteins or peptides of comparable size. Nonetheless, other works using SDS as dissociating
agent suggest that valuable information can be obtained.
The most recommended method for the study of cheese involves the stacking system of
Andrews () in alkaline gels containing 6M urea and the direct staining procedure of Blakesley
and Boezi, with Coomassie blue G250. Even so, it is reported that low molecular peptides can
be visualized using a silver staining technique involving glutaraldehyde as fixing agent, which
has not been applied yet to cheese analysis. On the other hand, the arrangement between
electrophoretic methods, or the application of a separation method followed by
electrophoresis, which is known as two dimensional electrophoresis, has been widely used in
the study of cheese ripening. Examples of this are works where SDS-PAGE is one dimension
and isoelectric focusing is the other, or where thin layer chromatography (TLC) is followed by
electrophoresis.
Usually, peptides resolved on PAGE are isolated and identified by excision of the bands or by
electroblotting. This last one is the most utilized, and peptides are identified by N-terminal
28
amino sequencing rather than mass spectroscopy (MS), due to they are stained. Nonetheless,
most of the major degradation products are known in most PAGE systems.
An alternative and potentially strong methodology to PAGE systems is capillary
electrophoresis (CE), which to date, is a technique not well implemented in the study of
cheese proteolysis, and counts with the capacity to resolve complicated mixtures of peptides,
using buffer filled capillary and an electric field that promotes separation based on the net
charge, molecular mass and Strokes’ radius.
Aroma of Cheddar cheese
Volatile flavor compounds identified in cheddar cheese includes a wide variety of acids,
alcohols, esters, aldehydes, ketones, phenolics and sulfur compounds summarized in the
following table.
Table 1 Aroma compounds of Cheddar cheese
COMPOUND
Dimethyl sulphide
Dimethyl disulphide
Dimethyl trisulphide
Hexanethiol
TYPE OF COMPOUND
Volatile sulfur compound
Volatile sulfur compound
Volatile sulfur compound
Volatile sulfur compound
Hydrogen sulfide
Methanethiol
Acetic acid
n-butanoic acid
Volatile sulfur compound
Volatile sulfur compound
Organic acid
Organic acid
n-decanoic acid
Isovaleric
acid
butanoic acid
Hexanoic acid
Butyric acid
3-methyl
n-octanoic acid
n-pentanoic acid
Phenyl acetic acid
Propionic acid
β - angelicalactone
γ - decalactone
δ - decalactone
δ - dodecalactone
6-(Z)-dodecenyl- γ - decalactone
Organic acid
Organic acid
Organic acid
Organic acid
Organic acid
Organic acid
Organic acid
Organic acid
Lactones
Lactones
Lactones
Lactones
Lactones
ATRIBUTE
Boiled cabbage
Cabbage, strong onion
Ripe cheese, garlic
burnt fat, sulfury meaty, fatty
garlic roasted burnt
Rotten egg
Rotten cabbage, fecal
Vinegar, sour, pungent
Sweaty, cheesy, fecal, rancid,
sharp
Rancid, waxy, soapy
Swiss cheese, waxy, sweaty, old
socks, fecal
Goat like
Sharp,
dairy-like,
cheesy,
buttery with a fruity nuance
Body odor, sweaty
Swiss cheese
Flowery
pungent
Coconut
Peachy, coconut
Cheesy, coconut
Soapy
29
sotolon
furaneol
Lactones
Lactones
Homofuraneol or ethyl furaneol
δ - octalactone
n-butanol
2-butanol
2,3 butanediol
p-cresol
ethanol
2 ethyl butanol
n-hexanol
Isobutanol
2-methyl-1-butanol
3-methyl-1-butanol
2 octanol
2,4 pentanediol
2 pentanol
2 phenyl ethanol
n-propanol
acetaldehyde
benzaldehyde
butanal
decanal
(E,E)-2,4-decadienal
Lactones
Lactones
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Alcohol
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
(E,Z)-2,4-decadienal
Aldehydes
Trans-4,5-epoxy-2-(E)-decenal
heptanal
(Z)-4-heptenal
n-hexanal
2-hexenal
isohexanal
2-methyl butanal
3-methyl butanal
2-methyl propanal
nonanal
(E)-2-nonenal
(Z)-2-nonenal
(E,Z)-2,6-nonadienal
(E,E)-2,4-nonadienal
octanal
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
pentanal
propanal
Phenyl aldehyde
propenal
Thiophen-2-aldehyde
acetone
acetophenone
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Aldehydes
Ketones
Ketones
Curry, seasoning
Sweet, caramel,burnt sugar,
strawberry
Caramel
Fruity, peachy, sweet
Floral, fruity, sweet
Alcoholic
Fruity
Unclean, medical, cowy, barny
Aalcohol
Fatty, green, floral
Wine
Fruity, alcohol, solvent, grainy
Mushroom, coconut, oil, rancid
Sweet, alcoholic, fruity, nutty
Rosy
Pungent
Sweet, pungent
Almond
Pungent
Soapy, flowery
Mayonnaise,
bread,
fatty,
tallow, fruity
Mayonnaise,
bread,
fatty,
tallow, fruity
Metallic
Fatty, oily, green
Creamy, biscuit
Green
Almond bitter, green, fatty
Dark chocolate, malt
Dark chocolate, malt
Malt
Green
Green, fatty
Green
Melon, cucumber
soapy
Green, fatty, soapy,
orange peel
Pungent, almond like
Solvent
Rosy
Solvent-like
Almond, musty, glue
fruity,
30
2-butanone
2,3 butanedione
(E)- β damascenone
2- heptanone
Ketones
Ketones
Ketones
Ketones
2-hexanone
1-hexen-3-one
3-hydroxy-2-butanone (acetoin)
3-methyl-2butanone
3-methyl-2-pentanone
2-nonanone
Ketones
Ketones
Ketones
Ketones
Ketones
Ketones
(Z)-1,5-octadien-3-one
2-octanone
Ketones
Ketones
1-octen-3one
Pentan-2one
2-tridecanone
2-undecanone
Ketones
Ketones
Ketones
Ketones
2-butyl acetate
n-butyl butyrate
n-butyl acetate
Ethyl acetate
Ethyl propionate
Ethyl butyrate
Ethyl hexanoate
Ethyl octanonoate
Methyl acetate
Methyl propionate
Methyl hexanoate
Propyl acetate
n-propyl butyrate
Geosmin
Guaicol
indole
limonene
linalool
α- pinene
Pyrazine, 2 acetyl
Pyrazine, 2-isobutyl-3-methoxy
Pyrazine,
2-isopropiyl-3methoxy
Pyrroline, 2 –acetyl-1
skatole
Thiazoline, 2-acetyl-2
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Esters
Terpene
Terpene
Pyrazine
Pyrazine
Pyrazine
pyrrol
(B. C. Weimer 2007; T. K Singh, Drake, and Cadwallader 2003)
Etheric
Buttery
Apple sauce
Blue cheese, fruity,
soapy
Fruity, ketone
Cooked vegetable,
Buttery
Camphor
musty,
Green, earthy, blue cheese,
fatty, musty, varnish
Green metallic
Floral, fruity, soapy, ketone,
musty
Mushroom
Acetone, sweet, fruity, ketone
Floral, fruity, green, musty,
tallow
Pear
Fruity, solvent, sweet
Fruity
Bubble gum, fruity
Fruity
Fruity
pineapple
Pineapple,
Earthy, moistened soil
Smoky, spicy
Mothball
Citrus
Sweet, floral, honey
Pine
Popcorn
Bell pepper
Earthy, soil, green, beany
Roasted
Unclean, mothball, fecal
roasted
31
REFERENCES
Alichanidis, Efstathios, Julia H. M. Wrathall, and Anthony T. Andrews. 1986. “Heat Stability
of Plasmin (milk Proteinase) and Plasminogen.” Journal of Dairy Research V53 (02):
pp. 259–269.
Alting, A. C., W. Engels, S. van Schalkwijk, and F. A. Exterkate. 1995. “Purification and
Characterization of Cystathionine (beta)-Lyase from Lactococcus Lactis Subsp.
Cremoris B78 and Its Possible Role in Flavor Development in Cheese.” Applied and
Environmental Microbiology V61 (11) (November 1): pp. 4037–4042.
Amarita, F., T. Requena, G. Taborda, L. Amigo, and C. Pelaez. 2001. “Lactobacillus Casei
and Lactobacillus Plantarum Initiate Catabolism of Methionine by Transamination.”
Journal of Applied Microbiology V90 (6): pp. 971–978.
Andersen, Lene T., Ylva Ardö, and Wender L.P. Bredie. 2010. “Study of Taste-active
Compounds in the Water-soluble Extract of Mature Cheddar Cheese.” International
Dairy Journal V20 (8) (August): pp. 528–536.
Anderson, D.L., V.V. Mistry, R.L. Brandsma, and K.A. Baldwin. 1993. “Reduced Fat
Cheddar Cheese from Condensed Milk. 1. Manufacture, Composition, and Ripening.”
Journal of Dairy Science V76 (10) (October): pp. 2832–2844.
Ardö, Y., and Frederiksberg. 1999. “Evaluating Proteolysis by Analysing the N Content of
Cheese Fractions.” Bulletin - International Dairy Federation V337: pp. 4–9.
Arora, Gulshan, Francois Cormier, and Byong Lee. 1995. “Analysis of Odor-Active Volatiles
in Cheddar Cheese Headspace by Multidimensional GC/MS/Sniffing.” J. Agric. Food
Chem. V43 (3): pp. 748–752.
Aston, J. W., and L. K. Creamer. 1986. “Contribution of the components of the water-soluble
fraction to the flavour of Cheddar cheese.” New Zealand Journal of Dairy Science and
Technology V21 (3): pp. 229–248.
Aston, J. W., and John R. Dulley. 1982. “Cheddar cheese flavor review.” Australian Journal
of Dairy Technology V37 (2) (June): pp. 59–64.
Avsar, Y.K., Y. Karagul-Yuceer, M.A. Drake, T.K. Singh, Y. Yoon, and K.R. Cadwallader.
2004. “Characterization of Nutty Flavor in Cheddar Cheese.” Journal of Dairy
Science V87 (7) (July): pp. 1999–2010.
Ayad, Eman H.E, Annette Verheul, Catrienus de Jong, Jan T.M Wouters, and Gerrit Smit.
1999. “Flavour Forming Abilities and Amino Acid Requirements of Lactococcus
Lactis Strains Isolated from Artisanal and Non-dairy Origin.” International Dairy
Journal V9 (10) (October): pp. 725–735.
Baer, A., and I. Ryba. 1992. “Serological Identification of Propionibacteria in Milk and
Cheese Samples.” International Dairy Journal V2 (5): pp. 299–310.
Balcão, Victor M., and F.Xavier Malcata. 1998. “Lipase Catalyzed Modification of Milkfat.”
Biotechnology Advances V16 (2) (March): pp. 309–341.
Bansal, N, Paolo Piraino, and P. L. H. McSweeney. 2009. “Determination of Proteolysis in
Cheese.” In Handbook of Dairy Foods Analysis, pp. 405–417. CRC Press.
Bassett, Harold J., and W.L. Slatter. 1953. “The Heat Resistance of Some Lactobacilli Found
in Swiss Cheese.” Journal of Dairy Science V36 (4) (April): pp. 409–413.
Bastian, E.D., C.G. Lo, and K.M.M. David. 1997. “Plasminogen Activation in Cheese Milk:
Influence on Swiss Cheese Ripening.” Journal of Dairy Science 80 (2) (February):
245–251.
32
Bastian, Eric D., and Rodney J. Brown. 1996. “Plasmin in Milk and Dairy Products: An
Update.” International Dairy Journal V6 (5) (May): pp. 435–457.
Benfeldt, Connie, and John Sørensen. 2001. “Heat Treatment of Cheese Milk: Effect on
Proteolysis During Cheese Ripening.” International Dairy Journal V11 (4–7) (July
11): pp. 567–574.
Benfeldt, Connie, John Sørensen, Katrine H. Ellegård, and Torben E. Petersen. 1997. “Heat
Treatment of Cheese Milk: Effect on Plasmin Activity and Proteolysis During Cheese
Ripening.” International Dairy Journal 7 (11) (October 24): 723–731.
Benkler, K. F., and G. A. Reineccius. 1980. “FLAVOR ISOLATION FROM FATTY FOODS
VIA SOLVENT EXTRACTION AND MEMBRANE DIALYSIS.” Journal of Food
Science V45 (4): pp. 1084–1085.
Bentley, Ronald, and Thomas G. Chasteen. 2004. “Environmental VOSCs––formation and
Degradation of Dimethyl Sulfide, Methanethiol and Related Materials.” Chemosphere
V55 (3) (April): pp. 291–317.
Beresford, Tom P, Nora A Fitzsimons, Noelle L Brennan, and Tim M Cogan. 2001. “Recent
Advances in Cheese Microbiology.” International Dairy Journal V11 (4–7) (July 11):
pp. 259–274.
Beuvier, E., and S. Buchin. 2004. “Raw Milk Cheeses.” In Cheese Chemistry, Physics and
Microbiology, Volume 1:319–345. Academic Press.
Bills, D.D., and E.A. Day. 1964. “Determination of the Major Free Fatty Acids of Cheddar
Cheese.” Journal of Dairy Science V47 (7) (July): pp. 733–738
Burbank, Helen M., and Michael C. Qian. 2005. “Volatile Sulfur Compounds in Cheddar
Cheese Determined by Headspace Solid-phase Microextraction and Gas
Chromatograph-pulsed Flame Photometric Detection.” Journal of Chromatography A
V1066 (1–2) (February 25): pp. 149–157.
Burbank, Helen, and Michael C. Qian. 2008. “Development of Volatile Sulfur Compounds in
Heat-shocked and Pasteurized Milk Cheese.” International Dairy Journal V18 (8)
(August): pp. 811–818.
Burton, H. 1986. “Microbiological aspects.” (200): 9–14.
Careri, M, and A Mangia. 2003. “Analysis of Food Proteins and Peptides by Chromatography
and Mass Spectrometry.” Journal of Chromatography A V1000 (1–2) (June 6): pp.
609–635.
Carunchia Whetstine, M E, P J Luck, M A Drake, E A Foegeding, P D Gerard, and D M
Barbano. 2007. “Characterization of Flavor and Texture Development Within Large
(291 Kg) Blocks of Cheddar Cheese.” Journal of Dairy Science V90 (7) (July): pp.
3091–3109.
Carunchia Whetstine, Mary E., Keith R. Cadwallader, and MaryAnne Drake. 2005.
“Characterization of Aroma Compounds Responsible for the Rosy/Floral Flavor in
Cheddar Cheese.” J. Agric. Food Chem. V53 (8): pp. 3126–3132.
del Castillo-Lozano, M. López, S. Mansour, R. Tâche, P. Bonnarme, and S. Landaud. 2008.
“The Effect of Cysteine on Production of Volatile Sulphur Compounds by Cheeseripening Bacteria.” International Journal of Food Microbiology V 122 (3) (March
20): pp. 321–327.
Chaintreau, Alain. 2001. “Simultaneous Distillation–extraction: From Birth to Maturity—
review.” Flavour and Fragrance Journal V16 (2): pp. 136–148.
Chatfield, Chris, and A. Collins. 1981. “Finding New Underlying Variables.” In Introduction
to Multivariate Analysis, pp. 57–78. CRC Press.
33
Chavarri, Felisa, Mailo Virto, Celia Martin, Ana I. Nájera, Arantza Santisteban, Luis J. R.
Barrón, and Mertxe De Renobales. 1997. “Determination of Free Fatty Acids in
Cheese: Comparison of Two Analytical Methods.” Journal of Dairy Research V64
(3): pp. 445–452.
Chich, Jean-François, Karine Marchesseau, and Jean-Claude Gripon. 1997. “Intracellular
Esterase from Lactococcus Lactis Subsp. Lactis NCDO 763: Purification and
Characterization.” International Dairy Journal V7 (2-3) (February): pp. 169–174.
Chopin, Alain. 1993. “Organization and Regulation of Genes for Amino Acid Biosynthesis in
Lactic Acid Bacteria.” FEMS Microbiology Reviews V12 (1–3) (September): pp. 21–
37.
Christensen J.E., Johnson M.E., and Steele J.L. 1995. “Production of Cheddar Cheese Using a
Lactococcus Lactis Ssp. Cremoris SK11 Derivative with Enhanced Aminopeptidase
Activity.” International Dairy Journal V5 (4): pp. 367–379.
Christensen, Jeffrey E., Edward G. Dudley, Jeffrey A. Pederson, and James L. Steele. 1999.
“Peptidases and Amino Acid Catabolism in Lactic Acid Bacteria.” Antonie Van
Leeuwenhoek V76 (1): pp. 217–246.
Christensen, K.R., and G.A. Reineccius. 1992. “Gas Chromatographic Analysis of Volatile
Sulfur Compounds from Heated Milk Using Static Headspace Sampling.” Journal of
Dairy Science V75 (8) (August): pp. 2098–2104
Christensen, Søren, Tove Wiegers, John Hermansen, and Lars Sottrup-Jensen. 1995. “Plasmaderived Protease Inhibitors in Bovine Milk.” International Dairy Journal V5 (5): pp.
439–449.
Christensen, T. M. I. E., A.-M. Bech, and H. Werner. 1991. “Methods for crude fractionation
(extraction and precipitation) of nitrogen components in cheese.” Bulletin International Dairy Federation (261): pp. 4–9.
Christensen, Tove M. I. E., Kristian R. Kristiansen, and Jesper S. Madsen. 1989. “Proteolysis
in Cheese Investigated by High Performance Liquid Chromatography.” Journal of
Dairy Research V56 (5): pp. 823–828.
Cliffe, A.J., D. Revell, and B.A. Law. 1989. “A Method for the Reverse Phase High
Performance Liquid Chromatography of Peptides from Cheddar Cheese.” Food
Chemistry 34 (2): 147–160.
Cliffe, Allan J., Jonathan D. Marks, and Francis Mulholland. 1993. “Isolation and
Characterization of Non-volatile Flavours from Cheese: Peptide Profile of Flavour
Fractions from Cheddar Cheese, Determined by Reverse-phase High-performance
Liquid Chromatography.” International Dairy Journal V3 (4–6): pp 379–387.
Collin, J. C., D. le Bars, I. Ryba, and A. Bear. 1990. “Comparative measurements of the
plasminogen-plasmin system of milk.”International Dairy Federation.
Collins, Y.F., P.L.H. McSweeney, and M.G. Wilkinson. 2004. “Lipolysis and Catabolism of
Fatty Acids in Cheese.” In Cheese Chemistry, Physics and Microbiology, Volume
1:373–389. Academic Press.
Collins, Yvonne F., Paul L.H. McSweeney, and Martin G. Wilkinson. 2003. “Lipolysis and
Free Fatty Acid Catabolism in Cheese: a Review of Current Knowledge.”
International Dairy Journal V13 (11): pp. 841–866.
Considine, T., A. Healy, A.L. Kelly, and P.L.H. McSweeney. 1999. “Proteolytic Specificity of
Elastase on Bovine Β-casein.” Food Chemistry V66 (4) (September): pp. 463–470.
Considine, T., Á. Healy, A.L. Kelly, and P.L.H. McSweeney. 2000. “Proteolytic Specificity of
Elastase on Bovine Αs1-casein.” Food Chemistry V69 (1) (April): pp. 19–26.
34
Considine, T., Á. Healy, A.L. Kelly, and P.L.H. McSweeney. 2004. “Hydrolysis of Bovine
Caseins by Cathepsin B, a Cysteine Proteinase Indigenous to Milk.” International
Dairy Journal V14 (2) (February): pp. 117–124.
Crow, V.L., T. Coolbear, P.K. Gopal, F.G. Martley, L.L. McKay, and H. Riepe. 1995. “The
Role of Autolysis of Lactic Acid Bacteria in the Ripening of Cheese.” International
Dairy Journal V5 (8): pp. 855–875.
Crow, Vaughan, Brian Curry, and Michelle Hayes. 2001. “The Ecology of Non-starter Lactic
Acid Bacteria (NSLAB) and Their Use as Adjuncts in New Zealand Cheddar.”
International Dairy Journal V11 (4–7) (July 11): pp 275–283.
Cserháti, Tibor. 2002. “Mass Spectrometric Detection in Chromatography. Trends and
Perspectives.” Biomedical Chromatography V16 (5): pp. 303–310.
Curioni, P.M.G., and J.O. Bosset. 2002. “Key Odorants in Various Cheese Types as
Determined by Gas Chromatography-olfactometry.” International Dairy Journal v12
(12): pp. 959–984. doi:10.1016/S0958-6946(02)00124-3.
Datta, N., A. J. Elliott, M. L. Perkins, and H. C. Deeth. 2002. “Ultra-high-temperature (UHT)
Treatment of Milk: Comparison of Direct and Indirect Modes of Heating” V57 (3):
pp. 211–227.
Deeth, H. C, and V. Touch. 2000. “Methods for detecting lipase activity in milk and milk
products.” Australian Journal of Dairy Technology V55 (3): pp. 153–168.
Dias, Benjamin, and Bart Weimer. 1998. “Conversion of Methionine to Thiols by Lactococci,
Lactobacilli, and Brevibacteria.” Applied and Environmental Microbiology V64 (9)
(September 1): pp. 3320–3326.
Dimos, Tony. 1992. “A comparative study by GC/MS of the flavour volatiles produced during
the maturation of full-fat and low-fat Cheddar cheese.”
Dirinck, P, and A De Winne. 1999. “Flavour Characterisation and Classification of Cheeses
by Gas Chromatographic–mass Spectrometric Profiling.” Journal of Chromatography
A V847 (1-2) (June 25): pp. 203–208.
Dobric, Nada, Gäetan K.Y. Limsowtin, Alan J. Hillier, Nicholas P.B. Dudman, and Barrie E.
Davidson. 2000. “Identification and Characterization of a Cystathionine Β/γ-lyase
from Lactococcus Lactis Ssp. Cremoris MG13631.” FEMS Microbiology Letters
V182 (2): pp. 249–254.
Drake, M.A., T.D. Boylston, K.D. Spence, and B.G. Swanson. 1996. “Chemical and Sensory
Effects of a Lactobacillus Adjunct in Cheddar Cheese.” Food Research International
V29 (3–4) (April): pp. 381–387.
El Soda, M., S.A. Madkor, and P.S. Tong. 2000. “Adjunct Cultures: Recent Developments
and Potential Significance to the Cheese Industry.” Journal of Dairy Science V83 (4)
(April): pp. 609–619.
El Soda, Morsi A. 1993. “The Role of Lactic Acid Bacteria in Accelerated Cheese Ripening.”
FEMS Microbiology Reviews V12 (1–3) (September): pp. 239–251.
El-Soda, M., S. A. Madkor, and P. S. Tong. 2000. “Evaluation of commercial adjuncts for use
in cheese ripening: 4. Comparison between attenuated and not attenuated lactobacilli.”
Milchwissenschaft V55 (5): pp. 260–263.
Engel, Wolfgang, Wolfgang Bahr, and P. Schieberle. 1999. “Solvent Assisted Flavour
Evaporation – a New and Versatile Technique for the Careful and Direct Isolation of
Aroma Compounds from Complex Food Matrices.” European Food Research and
Technology V209 (3): pp. 237–241.
35
Engels, W.J.M., R. Dekker, C. de Jong, R. Neeter, and S. Visser. 1997. “A Comparative Study
of Volatile Compounds in the Water-soluble Fraction of Various Types of Ripened
Cheese.” International Dairy Journal V7 (4) (April): pp. 255–263.
Exterkate, Fred A.., Arno C.. Alting, and Charles J. Slangen. 1995. “Conversion of αs1Casein-(24-199)-Fragment and β-Casein Under Cheese Conditions by Chymosin and
Starter Peptidases.” Systematic and Applied Microbiology V18 (1): pp. 7–12.
Fang, Yu, and Michael C. Qian. 2005. “Sensitive Quantification of Sulfur Compounds in
Wine by Headspace Solid-phase Microextraction Technique.” Journal of
Chromatography A V1080 (2) (July 8): pp. 177–185.
Farkye, Nana Y., and Patrick F. Fox. 1990. “Observations on Plasmin Activity in Cheese.”
Journal of Dairy Research V57 (3): pp. 413–418.
FDA. 2011. “Code of Federal Regulations Title 21.”
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=133.111.
Fennema, Owen R., and Damodaran. 1996. “Amino Acids, Peptides, and Proteins.” In Food
Chemistry, pp. 322–425. CRC Press.
Fernández, Manuela, Tanoj K. Singh, and Patrick F. Fox. 1998. “Isolation and Identification
of Peptides from the Diafiltration Permeate of the Water-Soluble Fraction of Cheddar
Cheese.” Journal of Agricultural and Food Chemistry V46 (11) (November 1): pp.
4512–4517.
Fernández, María, Wim van Doesburg, Ger A. M. Rutten, Joey D. Marugg, Arno C. Alting,
Richard van Kranenburg, and Oscar P. Kuipers. 2000. “Molecular and Functional
Analyses of themetC Gene of Lactococcus Lactis, Encoding Cystathionine β-Lyase.”
Applied and Environmental Microbiology V66 (1) (January 1): pp. 42–48.
Fernández-García, Estrella. 1996. “Use of Headspace Sampling in the Quantitative Analysis
of Artisanal Spanish Cheese Aroma.” Journal of Agricultural and Food Chemistry
V44 (7) (January 1): pp. 1833–1839.
Ferreira, A.D., and B.C. Viljoen. 2003. “Yeasts as Adjunct Starters in Matured Cheddar
Cheese.” International Journal of Food Microbiology V86 (1–2) (September 1): pp.
131–140.
Forde, Amanda, and Gerald F Fitzgerald. 2000. “Biotechnological Approaches to the
Understanding and Improvement of Mature Cheese Flavour.” Current Opinion in
Biotechnology V11 (5) (October 1): pp. 484–489.
Fox, P. F., T.P. Guinee, T. M. Cogan, and P. L. H. McSweeney. 2000. “Enzymatic
Coagulation of Milk.” In Fundamentals of Cheese Science, pp. 98135. Springer.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Fox, P. F., J. A. Lucey, and T. M. Cogan. 1990. “Glycolysis and Related Reactions During
Cheese Manufacture and Ripening.” Critical Reviews in Food Science and Nutrition
V29 (4) (January): pp. 237–253.
Fox, P. F., and P. L. H. McSweeney. 1996. “Proteolysis in Cheese During Ripening.” Food
Reviews International V12 (4): pp. 457–509.
Fox, P. F., and P. L. H. McSweeney. 1998. “Milk Lipids.” In Dairy Chemistry and
Biochemistry, pp. 67–141. Springer.
Fox, P. F., J. M. Wallace, S. Morgan, C. M. Lynch, E. J. Niland, and J. Tobin. 1996.
“Acceleration of Cheese Ripening.” Antonie Van Leeuwenhoek 70 (2): 271–297.
Fox, P.F. 1989. “Proteolysis During Cheese Manufacture and Ripening.” Journal of Dairy
Science V72 (6) (June): pp. 1379–1400.
36
Fox, P.F., and P.L.H. McSweeney. 2004. “Cheese: An Overview.” In Cheese Chemistry,
Physics and Microbiology, Volume 1:1–18. Academic Press.
Fox, P.F., and L. Stepaniak. 1993. “Enzymes in Cheese Technology.” International Dairy
Journal V3 (4–6): pp. 509–530.
Fox, PF, PLH McSweeney, and CM Lynch. 1998. “Significance of Non-starter Lactic Acid
Bacteria in Cheddar Cheese.” Australian Journal of Dairy Technology V53 (2): pp. 7–
8.
Gao, Song, Ed S. Mooberry, and James L. Steele. 1998. “Use of 13C Nuclear Magnetic
Resonance and Gas Chromatography To Examine Methionine Catabolism by
Lactococci.” Applied and Environmental Microbiology V64 (12) (December 1): pp.
4670–4675.
Gomez, Maria J., Sonia Garde, Pilar Gaya, Margarita Medina, and Manuel Nuñez. 1997.
“Relationship Between Level of Hydrophobic Peptides and Bitterness in Cheese Made
from Pasteurized and Raw Milk.” Journal of Dairy Research V64 (2): pp. 289–297.
Grandison, Alistair S., and Graeme D. Ford. 1986. “Effects of Variations in Somatic Cell
Count on the Rennet Coagulation Properties of Milk and on the Yield, Composition
and Quality of Cheddar Cheese.” Journal of Dairy Research V53 (4): pp. 645–655.
Grappin, Rémy, and Eric Beuvier. 1997. “Possible Implications of Milk Pasteurization on the
Manufacture and Sensory Quality of Ripened Cheese.” International Dairy Journal
V7 (12) (November 21): pp. 751–761.
Green, Margaret L., and Donald J. Manning. 1982. “Development of Texture and Flavour in
Cheese and Other Fermented Products.” Journal of Dairy Research V49 (4): pp. 737–
748.
Grieve, Paul A., and Barry J. Kitchen. 1985. “Proteolysis in Milk: The Significance of
Proteinases Originating from Milk Leucocytes and a Comparison of the Action of
Leucocyte, Bacterial and Natural Milk Proteinases on Casein.” Journal of Dairy
Research 52 (1): 101–112.
Griffiths, M. W. 1986. “Use of milk enzymes as indices of heat treatment.” Journal of Food
Protection V49 (9): pp. 696–705.
Gripon, Jean-Claude, V Monnet, G Lambert, and M. J. Desmazeaud. 1991. “Microbial
enzymes in cheese ripening.” In Food enzymology, vol 1.
Guinee, T P. 2004. “Salting and the Role of Salt in Cheese.” International Journal of Dairy
Technology V57 (2-3): pp. 99–109.
Habibi‐Najafi, Mohammad B., Byong H. Lee, and Barry Law. 1996. “Bitterness in Cheese: A
Review.” Critical Reviews in Food Science and Nutrition V36 (5): pp. 397–411.
Heegaard, Christian W., Lone K. Rasmussen, and Peter A. Andreasen. 1994. “The
Plasminogen Activation System in Bovine Milk: Differential Localization of Tissuetype Plasminogen Activator and Urokinase in Milk Fractions Is Caused by Binding to
Casein and Urokinase Receptor.” Biochimica Et Biophysica Acta (BBA) - Molecular
Cell Research V1222 (1) (May 26): pp. 45–55.
Hickey, D.K., K.N. Kilcawley, T.P. Beresford, and M.G. Wilkinson. 2007. “Lipolysis in
Cheddar Cheese Made from Raw, Thermized, and Pasteurized Milks.” Journal of
Dairy Science V90 (1) (January): pp. 47–56.
Holland, Ross, and Tim Coolbear. 1996. “Purification of Tributyrin Esterase from
Lactococcus Lactis Subsp. Cremoris E8.” Journal of Dairy Research V63 (1): pp.
131–140.
Home, D.S., and J.M. Banks. 2004. “Rennet-induced Coagulation of Milk.” In Cheese
Chemistry, Physics and Microbiology, Volume 1:pp. 47–70. Academic Press.
37
Huffman, L. M.;, and Kristoffersen. 1984. “Role of lactose in Cheddar cheese manufacturing
and ripening.” New Zealand Journal of Dairy Science and Technology V19 (2): pp.
151–162.
Hurley, M.J, B.M O’Driscoll, A.L Kelly, and P.L.H McSweeney. 1999. “Novel Assay for the
Determination of Residual Coagulant Activity in Cheese.” International Dairy
Journal V9 (8) (August): pp 553–558.
Hutton, J.T., and S. Patton. 1952. “The Origin of Sulfhydryl Groups in Milk Proteins and
Their Contributions to ‘Cooked’ Flavor.” Journal of Dairy Science V35 (8) (August):
pp. 699–705.
Jolly, Ramesh C., and Frank V. Kosikowski. 1975. “Quantification of Lactones in Ripening
Pasteurized Milk Blue Cheese Containing Added Microbial Lipases.” J. Agric. Food
Chem. V23 (6): pp. 1175–1176.
De Jong, Catrienus, and Herman T. Badings. 1990. “Determination of Free Fatty Acids in
Milk and Cheese Procedures for Extraction, Clean up, and Capillary Gas
Chromatographic Analysis.” Journal of High Resolution Chromatography V13 (2):
pp. 94–98.
K. Arfi, H.Spinnler, R. Tache, and P. bonnarme. 2002. “Production of Volatile Compounds by
Cheese-ripening Yeasts: Requirement for a Methanethiol Donor for S-methyl
Thioacetate Synthesis by Kluyveromyces Lactis.” Applied Microbiology and
Biotechnology V58 (4): pp. 503–510.
Knecht, E. 1999. “Lysomomal Cysteine Proteinases (2nd Edition).” Cell Biology International
V23 (8) (8) (August): pp 590–591.
Laan, Harry, Saw Eng Tan, Paul Bruinenberg, Gaëtan Limsowtin, and Malcolm Broome.
1998. “Aminopeptidase Activities of Starter and Non-Starter Lactic Acid Bacteria
Under Simulated Cheddar Cheese Ripening Conditions.” International Dairy Journal
V8 (4) (April): pp. 267–274.
Lane, C.N., and P.F. Fox. 1996 “Contribution of Starter and Adjunct Lactobacilli to
Proteolysis in Cheddar Cheese During Ripening.” International Dairy Journal V6 (7)
(July): pp. 715–728.
Lane, Cáit N., and Patrick F. Fox. 1997. “Role of Starter Enzymes During Ripening of
Cheddar Cheese Made from Pasteurized Milk Under Controlled Microbiological
Conditions.” International Dairy Journal V7 (1) (January): pp. 55–63.
Larsen, Lotte B., Connie Benfeldt, Lone K. Rasmussen, and Torben E. Petersen. 1996.
“Bovine Milk Procathepsin D and Cathepsin D: Coagulation and Milk Protein
Degradation.” Journal of Dairy Research V63 (1): pp 119–130.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen. 1990. “Influence of
Pasteurization on Fat and Nitrogen Recoveries and Cheddar Cheese Yield.” Journal of
Dairy Science V73 (3) (March): pp. 561–570. doi:10.3168/jds.S0022-0302(90)787036.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen. 1991. “Influence of
Pasteurization of Milk on Protein Breakdown in Cheddar Cheese During Aging.”
Journal of Dairy Science V74 (3) (March): pp. 727–740.
Law, A. J. R., J. M. Banks, D. S. Horne, J. Leaver, and I. G. West. 1994. “Denaturation of the
whey proteins in heated milk and their incorporation into Cheddar cheese.”
Milchwissenschaft V49 (2): pp. 63–67.
Lee, Won-Jae, Dattatreya S. Banavara, Joanne E. Hughes, Jason K. Christiansen, James L.
Steele, Jeffery R. Broadbent, and Scott A. Rankin. 2007. “Role of Cystathionine βLyase in Catabolism of Amino Acids to Sulfur Volatiles by Genetic Variants of
38
Lactobacillus Helveticus CNRZ 32.” Applied and Environmental Microbiology V73
(9) (May 1): pp. 3034–3039.
Lin, J. C. C., and I. J. Jeon. 1985. “Headspace Gas Sampling/GC Method for the Quantitative
Analysis of Volatile Compounds in Cheese.” Journal of Food Science V50 (3): pp.
843–844.
Liu, Shao-Quan, Ross Holland, and Vaughan L Crow. 2001. “Purification and Properties of
Intracellular Esterases from Streptococcus Thermophilus.” International Dairy
Journal V11 (1–2) (January): pp. 27–35.
Lloyd, G. T., J. F. Horwood, and I. Barlow. 1980. “The effect of yoghurt culture YB on the
flavour and maturation of Cheddar cheese.” Australian Journal of Dairy Technology
V35 (4): pp. 137–139.
Lynch, C.M., P.L.H. McSweeney, P.F. Fox, T.M. Cogan, and F.D. Drinan. 1996.
“Manufacture of Cheddar Cheese with and Without Adjunct Lactobacilli Under
Controlled Microbiological Conditions.” International Dairy Journal V6 (8–9)
(August): pp. 851–867.
Lynch, C.M., D.D. Muir, J.M. Banks, P.L.H. McSweeney, and P.F. Fox. 1999. “Influence of
Adjunct Cultures of Lactobacillus Paracasei Ssp. Paracasei or Lactobacillus Plantarum
on Cheddar Cheese Ripening.” Journal of Dairy Science V82 (8) (August): pp. 1618–
1628.
Lynch, Joanna M., David M. Barbano, and J. Richard Fleming. 2002. “Determination of the
Total Nitrogen Content of Hard, Semihard, and Processed Cheese by the Kjeldahl
Method: Collaborative Study.” Journal of AOAC International V85 (2): pp. 445–455.
Madkor, S.A., P.S. Tong, and M. El Soda. 2000. “Ripening of Cheddar Cheese with Added
Attenuated Adjunct Cultures of Lactobacilli.” Journal of Dairy Science V83 (8)
(August): pp. 1684–1691.
Manning, D. J., Helen R. Chapman, and Zena D. Hosking. 1976. “The Production of Sulphur
Compounds in Cheddar Cheese and Their Significance in Flavour Development.”
Journal of Dairy Research V43 (2): pp. 313–320.
Manning, Donald J., and Carolyn Moore. 1979. “Headspace Analysis of Hard Cheeses.”
Journal of Dairy Research V46 (3): pp. 539–545.
McSweeney, P.L.H., P.F. Fox, J.A. Lucey, K.N. Jordan, and T.M. Cogan. 1993. “Contribution
of the Indigenous Microflora to the Maturation of Cheddar Cheese.” International
Dairy Journal V3 (7): pp. 613–634.
McSweeney, P.L.H., P.F. Fox, and N.F. Olson. 1995. “Proteolysis of Bovine Caseins by
Cathepsin D: Preliminary Observations and Comparison with Chymosin.”
International Dairy Journal V5 (4): pp. 321–336.
McSweeney, Paul L. H., Sylvie Pochet, Patrick F. Fox, and Aine Healy. 1994. “Partial
Identification of Peptides from the Water-Insoluble Fraction of Cheddar Cheese.”
Journal of Dairy Research V61 (4): pp. 587–590.
McSweeney, Paul L.H., and Maria José Sousa. 2000. “Biochemical Pathways for the
Production of Flavour Compounds in Cheeses During Ripening: A Review.” Le Lait
V80 (3): pp. 293–324.
Metwalli, Ali A.M, Harmen H.J de Jongh, and Martinus A.J.S van Boekel. 1998. “Heat
Inactivation of Bovine Plasmin.” International Dairy Journal V8 (1) (January 1): pp.
47–56.
Milo, C., and G. A. Reineccius. 1997. “Identification and Quantification of Potent Odorants in
Regular-Fat and Low-Fat Mild Cheddar Cheese.” Journal of Agricultural and Food
Chemistry V45 (9) (September 1): pp. 3590–3594.
39
Mistry, V.V., and J.-L. Maubois. 2004. “Application of Membrane Separation Technology to
Cheese Production.” In Cheese Chemistry, Physics and Microbiology, Volume 1:pp.
261–VII. Academic
Mistry, V.V., L.E. Metzger, and J.L. Maubois. 1996. “Use of Ultrafiltered Sweet Buttermilk in
the Manufacture of Reduced Fat Cheddar Cheese.” Journal of Dairy Science V79 (7)
(July): pp. 1137–1145
Moio, Luigi, Dominique Langlois, Patrick Etievant, and Francesco Addeo. 1993. “Powerful
Odorants in Bovine, Ovine, Caprine and Water Buffalo Milk Determined by Means of
Gas Chromatography–olfactometry.” Journal of Dairy Research V60 (02): pp. 215–
222.
Molimard, P., and H.E. Spinnler. 1996 “Review: Compounds Involved in the Flavor of
Surface Mold-Ripened Cheeses: Origins and Properties.” Journal of Dairy Science
V79 (2) (February): pp. 169–184.
Mooney, J.S, P.F Fox, A Healy, and J Leaver. 1998. “Identification of the Principal Waterinsoluble Peptides in Cheddar Cheese.” International Dairy Journal V8 (9)
(September): pp. 813–818.
Muir, D.D., J.M. Banks, and E.A. Hunter. 1996. “Sensory Properties of Cheddar Cheese:
Effect of Starter Type and Adjunct.” International Dairy Journal V6 (4) (April): pp.
407–423.
Mulvihill, Donald M., and Patrick F. Fox. 1979. “Proteolytic Specificity of Chymosin on
Bovine Αs1-Casein.” Journal of Dairy Research V46 (4): pp. 641–651.
Naughton, M. A., and F. Sanger. 1961. “Purification and Specificity of Pancreatic Elastase.”
Biochemical Journal V78 (1): pp. 156–163.
Niven, C. F., L. G. Buettner, and James B. Evans. 1954. “Thermal Tolerance Studies on the
Heterofermentative Lactobacilli That Cause Greening of Cured Meat Products.”
Applied Microbiology V2 (1) (January): pp. 26–29.
O’Keeffe, Arthur M., Patrick F. Fox, and Charles Daly. 1978. “Proteolysis in Cheddar Cheese:
Role of Coagulant and Starter Bacteria.” Journal of Dairy Research V45 (3): pp. 465–
477.
O’Keeffe, R. B., P. F. Fox, and C. Daly. 1976. “Contribution of Rennet and Starter Proteases
to Proteolysis in Cheddar Cheese.” Journal of Dairy Research 43 (01): 97–107.
O’Shea, B.A., T. Uniacke-Lowe, and P.F. Fox. 1996. “Objective Assessment of Cheddar
Cheese Quality.” International Dairy Journal V6 (11-12) (November): pp. 1135–
1147.
Olivecrona, T., S. Vilaró, G. Bengtsson-Olivecrona, and P. F Fox. 1992. “Indigenous enzyme
in milk. II. Lipases in milk.” In Advanced Dairy Chemistry 1, 292–310.
Payne, Martin J., and Rohan G. Kroll. 1991. “Methods for the Separation and Concentration
of Bacteria from Foods.” Trends in Food Science & Technology V2 (0): pp. 315–319.
Peterson, S.D., and R.T. Marshall. 1990. “Nonstarter Lactobacilli in Cheddar Cheese: A
Review.” Journal of Dairy Science V73 (6) (June): pp. 1395–1410.
Piraino, Paolo, Eugenio Parente, and Paul L. H. McSweeney. 2004. “Processing of
Chromatographic Data for Chemometric Analysis of Peptide Profiles from Cheese
Extracts:  A Novel Approach.” Journal of Agricultural and Food Chemistry 52 (23)
(November 1): 6904–6911.
Precetti, A.S., M.P. Oria, and S.S. Nielsen. 1997. “Presence in Bovine Milk of Two Protease
Inhibitors of the Plasmin System.” Journal of Dairy Science V80 (8) (August): pp.
1490–1496.
40
Preininger, Martin, and Werner Grosch. 1994. “Evaluation of Key Odorants of the Neutral
Volatiles of Emmentaler Cheese by the Calculation of Odour Activity Values.” LWT Food Science and Technology V27 (3) (June): pp. 237–244.
Pripp, A.H, Shakeel-Ur-Rehman, P.L.H McSweeney, and P.F Fox. 1999. “Multivariate
Statistical Analysis of Peptide Profiles and Free Amino Acids to Evaluate Effects of
Single-strain Starters on Proteolysis in Miniature Cheddar-type Cheeses.”
International Dairy Journal V9 (7) (July): pp. 473–479.
Pripp, Are Hugo, Leszek Stepaniak, and Terje Sørhaug. 2000. “Chemometrical Analysis of
Proteolytic Profiles During Cheese Ripening.” International Dairy Journal V10 (4):
pp. 249–253.
Puchades, R., L. Lemieux, and R.e. Simard. 1989. “Evolution of Free Amino Acids During
the Ripening of Cheddar Cheese Containing Added Lactobacilli Strains.” Journal of
Food Science V54 (4): pp. 885–888.
Qian, M., and G. Reineccius. 2002. “Identification of Aroma Compounds in ParmigianoReggiano Cheese by Gas Chromatography/Olfactometry.” Journal of Dairy Science
V85 (6) (June): pp. 1362–1369.
Rehman, Shakeel-Ur, J.M Banks, E.Y Brechany, D.D Muir, P.L.H McSweeney, and P.F Fox.
2000. “Influence of Ripening Temperature on the Volatiles Profile and Flavour of
Cheddar Cheese Made from Raw or Pasteurised Milk.” International Dairy Journal
V10 (1–2) (January): pp. 55–65.
Richardson, B. C., L. K. Creamer, K. N. Pearce, and R. E. Munford. 1974. “Comparative
Micelle Structure: II. Structure and Composition of Casein Micelles in Ovine and
Caprine Milk as Compared with Those in Bovine Milk.” Journal of Dairy Research
V41 (2): pp. 239–247.
Roy, Denis, Isabelle Mainville, and Francine Mondou. 1997. “Selective Enumeration and
Survival of Bifidobacteria in Fresh Cheese.” International Dairy Journal V7 (12)
(November 21): pp. 785–793.
Sablé, Sophie, and Gilles Cottenceau. 1999. “Current Knowledge of Soft Cheeses Flavor and
Related Compounds.” J. Agric. Food Chem. V47 (12): pp. 4825–4836.
Seefeldt, K.E., and B.C. Weimer. 2000. “Diversity of Sulfur Compound Production in Lactic
Acid Bacteria.” Journal of Dairy Science V83 (12) (December): pp. 2740–2746.
Shakeel-Ur-Rehman, J. M Banks, P. L. H Mcsweeney, and P. F Fox. 2000. “Effect of ripening
temperature on the growth and significance of non-starter lactic acid bacteria in
Cheddar cheese made from raw or pasteurised milk.” International dairy journal V10
(1-2): pp. 45–53.
Singh, T. K, M. A Drake, and K. R Cadwallader. 2003. “Flavor of Cheddar Cheese: A
Chemical and Sensory Perspective.” Comprehensive Reviews in Food Science and
Food Safety V2 (4) (October 1): pp. 166–189.
Singh, T.K., P.F. Fox, P. Højrup, and A. Healy. 1994. “A Scheme for the Fractionation of
Cheese Nitrogen and Identification of Principal Peptides.” International Dairy
Journal V4 (2): pp. 111–122.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1995. “Water-Soluble Peptides in Cheddar
Cheese: Isolation and Identification of Peptides in the Diafiltration Retentate of the
Water-Soluble Fraction.” Journal of Dairy Research V62 (4): pp. 629–640.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy 1997. “Isolation and Identification of Further
Peptides in the Diafiltration Retentate of the Water-soluble Fraction of Cheddar
Cheese.” Journal of Dairy Research V64 (03): pp. 433–443.
41
Smacchi, Emanuele, and Marco Gobbetti. 1998. “Purification and Characterization of
Cystathionine Γ-lyase from Lactobacillus Fermentum DT41.” FEMS Microbiology
Letters V166 (2): pp. 197–202.
Smit, Gerrit, Bart A. Smit, and Wim J.M. Engels. 2005. “Flavour Formation by Lactic Acid
Bacteria and Biochemical Flavour Profiling of Cheese Products.” FEMS Microbiology
Reviews V29 (3): pp. 591–610.
Smith, A. M., and S. Nakai. 1990. “Classification of cheese varieties by multivariate analysis
of HPLC profiles.” Canadian Institute of Food Science and Technology Journal V23
(1): pp. 53–58.
Sousa, M.J, Y Ardö, and P.L.H McSweeney. 2001. “Advances in the Study of Proteolysis
During Cheese Ripening.” International Dairy Journal V11 (4–7) (July 11): pp. 327–
345.
Stuart, Mark R., Lan Szu Chou, and Bart C. Weimer. 1999. “Influence of Carbohydrate
Starvation and Arginine on Culturability and Amino Acid Utilization of Lactococcus
Lactis Subsp. Lactis.” Applied and Environmental Microbiology V65 (2) (February
1): pp. 665–673.
Subramanian, A., W.J. Harper, and L.E. Rodriguez-Saona. 2009. “Cheddar Cheese
Classification Based on Flavor Quality Using a Novel Extraction Method and Fourier
Transform Infrared Spectroscopy.” Journal of Dairy Science V92 (1) (January): pp.
87–94.
Suriyaphan, Orasa, MaryAnne Drake, X. Q. Chen, and Keith R. Cadwallader. 2001.
“Characteristic Aroma Components of British Farmhouse Cheddar Cheese.” Journal
of Agricultural and Food Chemistry V49 (3) (March 1): pp. 1382–1387.
Tammam, J. D, A. G Williams, J. Noble, D. Lloyd, J. D Tammam, A. G Williams, J. Noble,
and D. Lloyd. 2000. “Amino acid fermentation in non‐starter Lactobacillus spp.
isolated from Cheddar cheese, Amino acid fermentation in non‐starter Lactobacillus
spp. isolated from Cheddar cheese.” Letters in Applied Microbiology, Letters in
Applied Microbiology V30 (5) (May 1): pp. 370–374.
Tanaka, Hidehiko, Nobuyoshi Esaki, and Kenji Soda. 1985. “A Versatile Bacterial Enzyme:
L-methionine Γ-lyase.” Enzyme and Microbial Technology V7 (11) (November): pp.
530–537.
Thomsen, D., and H. Stapelfeldt. 1990. “Yield and quality of Danbo 45+ cheese from milk
heat treated at temperatures between 72 and 95°C.” (44): 39.
Tsakalidou, E., and G. Kalantzopoulos. 1992. “Purification and Partial Characterization of an
Esterase from Lactococcus Lactis Ssp Lactis Strain ACA-DC 127.” Le Lait V72 (6):
pp. 533–543.
Turner, K. W, and T. D Thomas. 1980. “Lactose fermentation in Cheddar cheese and the
effect of salt.” New Zealand Journal of Dairy Science and Technology V15 (3): pp.
265–276.
Upadhyay, V.K., P.L.H. McSweeney, A.A.A. Magboul, and P.F. Fox. 2004. “Proteolysis in
Cheese During Ripening.” In Cheese Chemistry, Physics and Microbiology, Volume
1:pp. 391–433.
Urbach, G. 1995. “Contribution of Lactic Acid Bacteria to Flavour Compound Formation in
Dairy Products.” International Dairy Journal V5 (8): pp. 877–903.
Urbach, Gerda. 1993. “Relations Between Cheese Flavour and Chemical Composition.”
International Dairy Journal V3 (4–6): pp. 389–422.
USDA. 2011. Dairy Products Annual Sumary.
42
Vazquez-Landaverde, P.A., J.A. Torres, and M.C. Qian. 2006. “Quantification of Trace
Volatile Sulfur Compounds in Milk by Solid-Phase Microextraction and Gas
Chromatography–Pulsed Flame Photometric Detection.” Journal of Dairy Science
V89 (8) (August): pp. 2919–2927.
Verdi, Robert J., and David M. Barbano. 1991. “Properties of Proteases from Milk Somatic
Cells and Blood Leukocytes.” Journal of Dairy Science V74 (7) (July): pp. 2077–
2081.
Visser, S. ;Hup. 1983. “Bitter flavour in cheese, 2: model studies on the formation and
degradation of bitter peptides by proteolytic enzymes from calf rennet, starter cells
and starter cell fractions.” Netherlands Milk and Dairy Journal V37 (3): pp. 169–180.
Visser, Servaas. 1993. “Proteolytic Enzymes and Their Relation to Cheese Ripening and
Flavor: An Overview.” Journal of Dairy Science V76 (1) (January): pp. 329–350.
Voigt, Daniela D., François Chevalier, John A. Donaghy, Margaret F. Patterson, Michael C.
Qian, and Alan L. Kelly. 2012. “Effect of High-pressure Treatment of Milk for
Cheese Manufacture on Proteolysis, Lipolysis, Texture and Functionality of Cheddar
Cheese During Ripening.” Innovative Food Science & Emerging Technologies V13
(0) (January): pp. 23–30.
Wallace J.M., and Fox P.F. 1997. “Effect of Adding Free Amino Acids to Cheddar Cheese
Curd on Proteolysis, Flavour and Texture Development.” International Dairy Journal
V7 (2): pp. 157–167.
Walstra, P., T. J Geurts, A. Noomen, A. Jellema, and M. A. J. S. van Boekel. 1999. “Cheese
Varieties.” In Dairy technology: principles of milk properties and processes, pp 651–
708.
Walstra, Pieter, Jan T. M. Wouters, and Tom J. Geurts. 2006. Dairy Science And Technology.
CRC Press.
Weimer, B. C. 2007. Improving the flavour of cheese. CRC Press.
Weimer, Bart, Kimberly Seefeldt, and Benjamin Dias. 1999. “Sulfur Metabolism in Bacteria
Associated with Cheese.” Antonie Van Leeuwenhoek 76 (1): 247–261.
Wijesundera, C., and L. Drury. “Role of milk fat in production of cheddar cheese flavour
using a fat-substituted cheese model.” Australian journal of dairy technology V54 (1):
pp. 28–35.
Wilkinson, Martin G., Timothy P. Guinee, Daniel M. O’Callaghan, and Patrick F. Fox. 1994.
“Autolysis and Proteolysis in Different Strains of Starter Bacteria During Cheddar
Cheese Ripening.” Journal of Dairy Research V61 (2): pp. 249–262.
De Wit, M., G. Osthoff, B.C. Viljoen, and A. Hugo. 2005. “A Comparative Study of Lipolysis
and Proteolysis in Cheddar Cheese and Yeast-inoculated Cheddar Cheeses During
Ripening.” Enzyme and Microbial Technology V37 (6) (November 1): pp. 606–616.
de Wit, Rudy, and Hans Nieuwenhuijse. 2008. “Kinetic Modelling of the Formation of
Sulphur-containing Flavour Components During Heat-treatment of Milk.”
International Dairy Journal V18 (5) (May): pp. 539–547.
Wong, N. P, R. Ellis, and D. E LaCroix. “Quantitative determination of lactones in Cheddar
cheese.” Journal of Dairy Science V58 (10): pp. 1437–1441.
Yvon, Mireille, Claire Chabanet, and Jean-Pierre Pélissier. 1989. “Solubility of Peptides in
Trichloroacetic Acid (TCA) Solutions Hypothesis on the Precipitation Mechanism.”
International Journal of Peptide and Protein Research V34 (3): pp. 166–176.
Yvon, Mireille, and Liesbeth Rijnen. 2001. “Cheese Flavour Formation by Amino Acid
Catabolism.” International Dairy Journal V11 (4–7) (July 11): pp. 185–201.
43
Zehentbauer, Gerhard, and G. A. Reineccius. 2002. “Determination of Key Aroma
Components of Cheddar Cheese Using Dynamic Headspace Dilution Assay.” Flavour
and Fragrance Journal V17 (4): pp. 300–305
44
CHAPTER 2
EVALUATION OF THE EFFECT OF HEAT-SHOCK AND
PASTEURIZATION TREATMENTS OF CHEESE MILK ON RIPENING
RELATED CHANGES IN CHEDDAR CHEESE
Lemus Muñoz, Freddy Mauricio, Qian Michael
Department of Food Science and Technology, Oregon State University
Corvallis Oregon 97331
45
ABSTRACT
The effect of heat shock (66C for 30 sec) and pasteurization (72C for 15C) treatments of
cheese milk was studied during the ripening stage of Cheddar cheese. Proteolysis was
investigated by a fractionation scheme, resulting in an insoluble fraction analyzed by urea
polyacrylamide gel electrophoresis (Urea-PAGE), and a soluble fraction which was further
investigated through water soluble nitrogen (WSN), trichloroacetic acid soluble nitrogen
(TCA-SN) and phosphotungstic acid soluble nitrogen (PTA-SN) analyzed by total Kjeldahl
nitrogen content (TKN). Reversed phase high performance liquid chromatography (RPHPLC) was used to study the peptide profile of the water soluble fraction. Lipolysis was
studied by levels of individual free fatty acids determined through gas chromatography-flame
ionization detection (GC-FID) after isolation employing solid phase extraction (SPE). Volatile
sulfur compounds were studied using head space solid phase micro-extraction (SPME)
coupled with gas chromatography-pulsed flame photometric detection (PFPD).
The Urea-PAGE method was able to differentiate samples according their age, but it could not
discriminate samples regarding their treatment. Nonetheless, measurements of total Kjeldahl
Nitrogen (TKN) of the WSN, TCA-SN, and PTA-SN fractions, and the principal component
analysis of the RP-HPLC peptide profile of the WSN fraction, revealed differences in the rate
and pattern of proteolysis. Levels of total nitrogen for the WSN, TCA-SN and PTA-SN
fractions increased as cheese aged and were lower for samples made from pasteurized milk,
indicating that primary and secondary proteolysis were faster for samples made with heatshocked milk. It was obtained a PCA model with 3 principal components that accounted for
the 82.6% of the variability from data collected. This model discriminate the samples
according age and quality, suggesting the samples undergo more or faster proteolysis. FFA
profiles reveal minor but not significant difference in the extension of the inactivation of the
lipoprotein lipase and its role during ripening, which is related to a higher lipolytic activity for
heat-shocked samples. The Volatile Sulfur Compounds (VSC) analysis showed that cheeses
made from heat-shocked milk developed higher concentrations of H2S, DMS and MeSH,
suggesting slower catabolism of sulfur containing amino acids in cheese made with
pasteurized milk.
46
INTRODUCTION
Although cheeses made from raw milk develop stronger flavors and ripen faster (P. F. Fox et
al. 1999), because of the commercial interest in maintaining a microbiological safe product
during longer periods of time in the market, thermal treatments of cheese milk have become
the first unit operation that determines the characteristic flavor of commercial cheddar cheese
during the cheese making process. However, differences in manufacturing procedures, such as
type of heat treatment applied to the cheese milk, result in different flavor profiles,
corresponding to its evident influence on the development of attributes. Whereby, it is of
industrial interest to find a well adjusted way to reduce microbial number without significant
effects on organoleptic characteristics and/or nutritional components, while emerging
technologies can be totally implemented.
Comparisons between the difference in the development of attributes of cheddar cheese made
from raw milk and pasteurized milk have been rigorously studied (Lau, Barbano, and
Rasmussen 1990; Lau, Barbano, and Rasmussen 1991; Grappin and Beuvier 1997; Hickey et
al. 2007; Cáit N. Lane and Fox 1997). However, complementary analysis of temperature
effects on cheese milk during ripening are still required to fulfill the gap of knowledge in
differences of standard heat treatments.
Thus, its is worthy to focus again in common
approaches such as heat-shock (66 °C for 30 sec) and pasteurization (72 °C for 15 sec), which
are slightly different. Furthermore, it is key to maintain the scope in low processing
temperatures, based on that previous studies show that defects in body and flavor, and a
reduced proteolytic activity arise from severe heat treatments due to inactivation of Plasmin by
thiol-disulphide reactions with denatured-lactoglobulin and formation of complex between
caseins and b-lactoglobulin, that result in the interference of whey protein in the maturation
of cheese (Thomsen and Stapelfeldt 1990; Law et al. 1994)
Essentially, heat affects the indigenous enzyme activity and thermo-labile non-starter lactic
acid bacteria (NSLAB) of milk. On the one hand, the enzyme lipoprotein lipase (LPL) and its
activator (C-II apolipoprotein-glutamic acid) are practically inactivated due to their sensitivity
to temperature. Indeed, the 73-95% of LPL activity is lost after pasteurization of milk. On the
other hand, because of Plasmin and its pro-enzyme Plasminogen depend on a thermo-labile
heterogeneous system of inhibitors (Heegaard, Rasmussen, and Andreasen 1994; Metwalli, de
47
Jongh, and van Boekel 1998; E.D. Bastian, Lo, and David 1997; Precetti, Oria, and Nielsen
1997), heat treatment of cheese milk leads to higher plasmin acitvity during the ripening stage
(Grappin and Beuvier 1997). In addition, heat induces the destruction of some beneficial
NSLAB (Lau, Barbano, and Rasmussen 1991; Rehman et al. 2000), or results in a significant
reduction of NSLAB species depending on the severity of the treatment (P.L.H. McSweeney
et al. 1993; Roy, Mainville, and Mondou 1997; Beuvier and Buchin 2004), which has a direct
impact on the texture and quality of the product as consequence of the lactose metabolism, in
which a reduced racemization of L(+)-lactate and formation of calcium lactate crystals lactate
take place.
In the case of cheddar cheese and other hard and semi-hard varieties, standard heat treatments
of cheese milk do not influence secondary proteolysis, which is mainly depend on proteinases
and peptidases from LAB and other adjunct cultures, but affect primary proteolysis where the
main enzymatic activity is provided by the rennet and the indigenous milk proteinase plasmin
(P. F. Fox 1989; S. Visser 1993; P.L.H. McSweeney et al. 1993). However, rennet enzymatic
activity is not directly affected by heat treatments of cheese milk, but it is reported that the
accessibility to αs1 caseins is altered. On the contrary, proteolytic activity of plasmin is
affected directly by temperature (P. F. Fox 1989; Benfeldt et al. 1997). Previous studies
showed that cheese manufactured from pasteurized milk displayed higher proteolytic digestion
of β-casein and an analogous increase in the amount of γ-casein compared to cheeses
manufactured from raw milk (Grappin and Beuvier 1997). However, it has been reported as
well that in spite of the stability of Plasmin at elevated temperatures, its activity decrease as
the temperature and the holding time increased (Benfeldt et al. 1997; Benfeldt and Sørensen
2001).
Regarding the degradation of milk fat, it has been documented that the level of lipolysis in
cheese made from pasteurized milk was lower than that attained in cheese made from raw
milk (P.L.H. McSweeney et al. 1993; Shakeel-Ur-Rehman et al. 2000), which can be
explained by the inactivation of the indigenous enzyme LPL and differences in growth rates
of NSLAB (P.L.H. McSweeney et al. 1993; Roy, Mainville, and Mondou 1997; Beuvier and
Buchin 2004), resulting in final product with lower levels of individual free fatty acids and
48
other compounds like methyl ketones and lactones in comparison to those found in cheese
made from raw milk (Olivecrona et al. 1992).
Thus, in order to understand the influence of heat treatments of cheese milk on the
biochemical changes of cheddar cheese during ripening; Calculating levels of individual FFA
(Qian and Reineccius 2002), monitoring the development of volatile sulfur compounds (H. M.
Burbank and Qian 2005; H. Burbank and Qian 2008), and tracking proteolysis through the
nitrogen content of certain fractions (J. M. Lynch, Barbano, and Fleming 2002), and the
analysis of peptide profiles by chromatographic and electrophoretic techniques (Dirinck and
De Winne 1999; Benfeldt and Sørensen 2001; Sousa, Ardö, and McSweeney 2001) are
important and appropriate approaches to be used. The objective of this study was to track the
proteolysis and lipolysis during the ripening of cheddar cheese samples to investigate agerelated changes resulting from different heat treatment practices of cheese milk.
MATERIALS AND METHODS
CHEESE SAMPLES
Cheeses were manufactured in a local dairy plant, using pasteurized milk and heat – shocked
milk. For the pasteurization treatment, milk was heated at 72 C for 16 seconds and cheese was
made according with standard protocols. Milk for the heat-shocked treatment was heated at 60
C for 30 seconds and the cheese was made according with standard protocols. Three blocks of
cheese made from each milk treatment were selected randomly from three consecutive
manufacturing days. All cheeses were aged using the same conditions at manufacturer’s
facility, Every month a 2 lb portion was sampled from each block and sent to the lab, where
samples are stored at (-37C) to stop ageing process until analysis is completed.
FREE FATTY ACIDS ANALYSIS
Chemicals
Pentanoic acid, heptanoic acid, nonanoic acid, undecanoic acid, and heptadecanoic acid were
used as internal standards, they were purchased from Eastman (Rochester, N.Y., U.S.A).
Butanoic acid, hexanoic acid, octanoic acid, decanoic acid, dodecanoic acid, tetradecanoic
acid, 9-tetradecanoic acid, hexadecanoic acid, 9-hexadecanoic acid, octadecanoic acid, 9-
49
octadecanoic acid, 9,12-octadecanoic acid and 6,9,12 octadecanoic acid were used for the
standard stock solution, and were obtained from Aldrich Chemical Co. Inc (Milwaukee,
Wisconsin, U.S.A). Heptane, Isopropanol, Sulfuric acid, anhydrous sodium sulfate,
chloroform, formic acid and diethyl ether were obtained from Fisher.
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Six
grams of this previously freeze-ground cheese, 1 ml of 2N sulphuric acid and 1 ml of internal
standard solution (C5:0, C7:0, C9:0, C11:0 and C17:0 in 1:1 hetpane-isopropanol) were mixed
with 7 grams of anhydrous sodium sulfate and 20 ml of 1:1 diethyl ether- heptane in a 40 ml
amber vial using a sonicator and manual agitation. During sonication, the salt-slurry solution
is initially exposed for 15 minutes, after which each vial is shake vigorously to continue with a
second sonication period of 20 minutes. With a glass-Pasteur pipette, the sample extract
(solvent) is transferred to an AccuBOND amino cartridge (Agilent Technologies) conditioned
previously with 10 ml of heptane. After the addition of the sample, the column is washed with
5 ml of 2:1 Chloroform-Isopropanol to remove non volatile triglycerides and phospholipids
using a manifold vacuum chamber. Once the washing step is complete, free fatty acids are
eluted with 5ml of 2% formic acid in diethyl ether, collected in a 20 ml vial and stored in the
freezer until GC analysis.
Chromatography
The analysis was performed using a Hewlett-Packard 5890 gas chromatograph equipped with
a flame ionization detector (FID). Samples were analyzed on a DB-FFAP column (15m x
0.53mm ID, 1 m film thickness; Supelco Wax10, Supelco U.S.A). Injector and detector
temperatures were 250C. Nitrogen was used as carrier gas at a flow rate of 15 ml per minute at
a split ratio of 1 to 1. The oven temperature was programmed for a 2 minutes hold at 60C,
raised to 230C at a rate of 8C per minute with a hold of 20 minutes at 230C.
Quantitative analysis
The levels of free fatty acids concentrations were calculated based on individual peak area
from GC-FID response in comparison to the internal standard peak area, by using standard
50
calibration curve of individual free fatty acid using Peak Simple software (SRI instruments,
Torrance, CA). Each experimental value corresponds to the average of the 3 extraction
replicates.
VOLATILE SULFUR COMPOUNDS (VSC’S)
Chemicals
Dimethyl sulfide (DMS) was purchased from TCI America (Portland, OR, U.S.A.); gaseous
methanethiol (MeSH) was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wisconsin,
U.S.A), and a solution was prepared by bubbling the gas into cold methanol; a H2S solution
was prepared by dissolving Na2S.9 H2O (Sigma Co) in acidic water stabilized with citric acid
(pH 3).
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Then one
gram of this freshly prepared powder is added to a 20ml vial (formerly flushed with argon),
followed by the addition of 4 ml of 1M citric acid and 20 l of the internal standard solution.
After addition of sample vials were immediately sealed with screw caps with teflon-lined
silicone septa. The vials used in this study were previously deactivated with DMTCS 5%
solution in toluene, toluene, methanol and distillate water.
The volatile sulfur compounds were extracted with an 85 m Carbox-PDMS fiber (Supelco,
Bellefonte, PA, U.S.A.). Prior to use, the fiber was conditioned at 300 C for 90 minutes. The
fiber was then placed into a SPME adapter of a CombiPAL autosampler (CTC analytics AG,
Zwingen, Switzerland) Fitted with a vial heater/agitator. Samples were pre-equilibrated at 500
RPM at 40C for 5 minutes, and the extraction of VSC’s was done at 250 RPM at 40C for 25
minutes. The desorption time was 5 minutes and 30 seconds.
Chromatography
The analysis was performed using a Varian CP-3800 gas chromatograph (Varian, Walnut
Creek, CA, U.S.A.) equipped with a pulsed flame photometric detector (PFPD). The
separation of analytes was made using a DB-FFAP fused silica capillary column (30m, 0.32
51
mm ID and 1 m film thickness; Agilent, Palo Alto, CA, USA) and nitrogen as carrier gas at
constant flow at 2 ml per minute. The injector temperature was 300 C and it was in the
splitless mode. The oven temperature was programmed for a 3 minutes hold at 35C, raised to
150C at a rate of 10C per minute, held for 5 minute, and then heated to 220C at a rate of 20C
per minute with a final hold of 3 minutes. The PFPD was held at 300 C and 450 V with the
following flow rates: Air 1 at 17 ml per min, H2 at 14 ml per min, and Air 2 at 10 ml/min. The
detector response signal was integrated using the software Star Workstation 6.2, Varian)
Quantitative analysis
Matrix effect
In order to retain the matrix effect during the construction of the calibration curves, cheese
powder from the “youngest sample” is used. It is de-volatilized by exposure to room
conditions in a hood for 2 hours. Then 1 gram of powder is added to 4 ml of 1M citric acid in
a 20 ml vial and exposed to a 50C water bath for 30 mins, prior to the addition of standards
and internal standard solutions.
Sulfur standards and internal standard preparation
Two internal standards were used for the quantification of VSC’s: ethyl methyl sulfide (EMS)
for H2S, MeSH and DMS, and isopropyl disulfide (IsoProDS) for DMDS and DMTS. The
concentration of the internal standard solution was 500 ppm EMS and 500 ppm IsoProDS in
methanol. Calibration curves were constructed by spiking cheese samples with a range of
known concentrations of H2S, MeSH and DMS. Hydrogen sulfide (H2S) was prepared by
dissolving Na2S.9 H2O in acidic water (pH = 3). Different concentrations of sodium sulfide
solutions were made, and the concentrations of H2S were calculated based on the amounts of
salt added to the matrix. A standard solution of 100 ppm of DMS was individually prepared in
cooled methanol (-15C), and dilutions were made with cooled methanol at the same
temperature. The mesh standard was prepared as following: 1) newly deactivated, recently
flushed with argon, and cooled vials were used; 2) The original standard solution was made by
bubbling pure MeSH into cooled methanol; 3) Dilutions were prepared by taking aliquots
from the original solution contained in a sealed vial, through the teflon-lined silicone septa by
using a syringe. And then injecting the aliquots into new sealed vials containing proportional
amount of cooled methanol through the septa; 4) 1 gr of devolatilized cheese is added to a
52
recently flushed vial (argon was used), which is immediately flushed again; 5) simultaneous
argon flushing and addition of 4ml of “free” dissolved oxygen-1M citric acid solution and
quick sealing of the vials; 6) Addition of 20 l of internal standard and MeSH standard
through septa. The identification of target compounds was made by comparing retention times
with those of pure standards. Ratios of the square root of the standard area to the
corresponding square root of the internal standard area were plotted Vs concentration ratios to
determine the relation between the response and concentration for the unknowns. Triplicate
analysis was performed for all samples
PROTEOLYSIS
Chemicals
Sulfuric acid was purchased from Fisher Scientific International Inc. (Pittsburgh, PA, U.S.A.);
Trichloroacetic acid was purchased from Alfa Aesar (Royston, UK); and phosphotungstic acid
was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wis, U.S.A)
Sample preparation and fractionation
From each 2lb block of cheese, 60 grams are blended with 120 ml of distilled water pre-heated
to 55C. The mixture is blended for 5 minutes and the homogenate is incubated at 55C for 1
hour. Then the pH is adjusted to 4.6 with 1M HCl and the mixture is centrifuged at 3000g for
30 minutes at 4C. Suspension and supernatant were filtered thoroughly 3 times through glass
wool. The filtrate was safe at -20C for macro blog digestion method analysis, and RP-HPLC
analysis. The insoluble pellet was frozen at -20C for further Urea-PAGE gel electrophoresis
analysis.
The trichloroacetic acid soluble nitrogen fraction (TCA-SN) was prepared by the addition of
25 ml of pH 4.6 soluble fraction (WSN) to 25 ml of 24% trichloroacetic acid solution. Then
the mixture is equilibrated for 2 hours at room temperature and filtered through filter paper
Whatman No 40 before macro blog digestion method analysis.
For the phosphotungstic acid soluble nitrogen fraction (PTA-SN), 10 ml of WSN are added to
7 ml of 3.95 M H2SO4 and 3 ml of 33% phosphotungstic acid solution. Then the mixture is
53
equilibrated overnight at 4C and filtered through filter paper Whatman No 40 before macro
blog digestion method analysis.
Duplicate analysis was performed for all samples.
Macro blog digestion (Kjeldahl Digestion)
From the fractions collected an aliquot (2 ml for the Water soluble fraction, 1ml for TCA-SN
and 1 ml for PTA-SN) is added into a 70 ml Kjeldahl Digestion flask with 10 ml of H2SO4 and
the catalyst pellet containing 0,075 and 1,5 grams of mercuric oxide and potassium sulfate
respectively. The mixture is warmed to 150 C and hold for 1 hour, then heated to 250 C and
hold for 1 hour, and finally heated to 350 C and hold for 2 hours. After digestion the sample is
cooled down overnight to room temperature, and diluted with distillate water to 70 ml,
followed by a gentile agitation. Then a 5 ml aliquot is used to determine the nitrogen content
by a rapid flow analyzer FOSS II.
Reversed phase High performance liquid chromatography analysis
The RP-HPLC analysis was performed using a Shimadzu 6 series liquid chromatograph
(Shimadzu scientific instruments, Kyoto Japan), consisting of an autosampler, 2 pumps, a
multi-wavelength spectrophotometer and a controller unit. It was used a nucleosil RP-8
analytical column (250x 4mm, 5 m particle size, 300 A pore size) and a guard column (4.6
x10 mm) from waters (Milford, MA, U.S.A.). The mobile phase consists of solvent A (0.1%
TFA in deionized and vacuum filtered water) and solvent B (0.1% TFA in acetonitrile). The
elution was monitored at 214nm. The following gradient elution was performed: 1) 100%
solvent A for 5 minutes followed by a linear gradient to 55% solvent B (v/v); 2) elution at
55% solvent B for 6 minutes followed by a linear gradient to 60%; 3) elution at 60% solvent B
for 3 minutes; 4) The column is washed using 95% solvent B during 5 minutes; 5) the column
is equilibrated using 100% solvent A during 10 minutes. The sample (WSN fraction) was
dissolved in solvent A (10 mg per ml) and then micro-centrifuged at 14000 RPM for 10
minutes. An aliquot of 40 l from the extract was injected to a flow rate of 0.75 ml per min.
54
Electrophoresis
Samples of the water-insoluble nitrogen fraction were dry frozen prior to analysis. Samples
were dissolved in a buffer (0.75 g tris, (hydroxymethyl) methylamine, 49 gr urea and 0.4 ml
concentrated HCl, 0.7 ml 2-mercaptoethanol and 0.15 gr bromophenol blue, dissolved to
100ml) and hold at 50C for 40 min. Urea-polyacrylamide gel electrophoresis (urea-PAGE)
was carried out using a Protean II xi cell vertical slab unit (Bio-Rad Laboratories ltd., Hemel
Hempstead, Herts, UK). Urea-PAGE gels (12.5%) were prepared and run according to the
method or Ardö (1999). Reagents used were obtained by Sigma-Aldrich, Inc and Fisher
Scientific.
STATISTICAL ANALYSIS
A two-way analysis of variance (ANOVA) on data was carried out using a general linear
model procedure with Turkey’s pair wise comparison at 95% confidence level, using the
package Minitab 16 (minitab Ltda., Coventry, UK).
RESULTS AND DICUSSION
FREE FATTY ACIDS FFA
The levels of lipolysis obtained in this work, measured as the amount of individual FFA,
showed subtle but interesting difference between the cheeses prepared with pasteurized milk
and those made with heat-shocked milk. In spite of the initial loss of some short and medium
chain FFA during whey draining, it can be seen the tendency of FFA to increase during
maturation of samples. Indeed, it is evident in Figures (2) that short chain FFA tend to
increase faster, reaching concentrations about 4 times the initial one by the end of the
observation period for both type of treatments. On the one hand, this can be related to the
mobility and better access of enzymes to these substrates, which are essentially located at the
positions sn-1 and sn-3 (Balcão and Malcata 1998). On the other hand, this behavior indicates
that enzymatic activity is most likely dominated by lipases since they are specific for the outer
ester bonds of tri- or diacylglycerides (Deeth and Touch 2000). In addition, this last
observation is support by the fact that in spite of the low change, long chain FFA increased
their concentration during ripening, which is directly related to lipolytic activity rather than
activity promoted by esterases.
55
Despite the most dominant peaks or FFA in the chromatograms correspond to C14:0, C16:0,
C18:0 and C18:1, due to their significantly lower odor thresholds (Molimard and Spinnler
1996), they are not considered as important contributors to the overall aroma of Cheddar
cheese. Contrary, in the case of FFA such as C4:0, C6:0, C8:0, C10:0 and C12:0, it is known
that in spite of their lower concentration, they contribute directly and indirectly to the
characteristic aroma of Cheddar cheese, and it was seen that their rate of generation depend on
the type of heating treatment of cheese milk. Nonetheless, no significant difference was found,
but it is evident the trend of cheeses prepared with heat shocked milk to develop slightly
higher levels of short chain FFA. This observation can be related to higher LPL activity in
samples made with heat-shocked milk, since it is well known that 78 C for 10 seconds are
required for the complete inactivation of this enzyme (Driessen, 1989), and it has been
accepted that despite 72 C for 15 seconds inactivate the enzyme extensively, it still contributes
to lipolysis in pasteurized milk cheese. Therefore, a lower temperature treatment is expected to
affect the activity of this enzyme to a lesser extent. Indeed, this can be supported by the fact
that LPL is specific for the sn-1 and sn-3 positions of mono, di and triacylglicerides
(Olivecrona et al. 1992), which are the positions where C4:0 and C6:0 are predominately
located along with other unsaturated FFA (Balcão and Malcata 1998). However, since the
findings are subtle differences rather than significant, it is important to keep in mind that this
kind of ripening behavior might not be reproducible.
Alternatively, based on the fact that the principal lipolytic activity is provided by LAB
enzymes, which are mostly intracellular and have nothing to do with the heat treatment of
milk, it could be thought that the differences found in this work can be attributed to physical
alteration of milk as consequence of the heating treatment of cheese milk (Dalgleish and
Banks., 1991), resulting in a more difficult access for lipases and esterases of LAB in
pasteurized milk cheeses to their substrates (Hickey et al. 2007), and probably higher LPL
activity in Heat-shocked milk cheeses.
56
Figure 1 FFA chromatogram Heat –schokecd Vs Pasteurized milk
Figure 2 Development of individual FFA in Heat-shocked and Pasteurized milk cheese
57
58
Volatile Sulfur Compounds VSC’s
The decomposition of sulfur containing amino acids (cysteine and methionine) is another
biochemical event occurring during Cheddar cheese ripening resulting in the characteristic
aroma of this variety. In addition to free fatty acids (FFA), Volatile sulfur compounds (VSC’s)
correlate with good Cheddar cheese flavor (B. Weimer, Seefeldt, and Dias 1999). When
smelled alone they smell like garlic, onion, cabbage and skunk, but when they are mixed, they
contribute to pleasant Cheddar cheese flavor notes. It has been reported that high
concentrations of H2S, MeSH, and DMS are found in Cheddar Cheese, while DMDS, DMTS
and 3-methylthiopropionaldehyde have low concentrations. Other compounds such as
Carbonyl sulfide, carbon disulfide and dimethyl sulfone are not important contributors (H. M.
Burbank and Qian 2005).
59
The results from this work suggest that after assuring the stability of samples through the use
of an organic acid buffer solution (citric acid 1M), and the proper de-activation of injection
liner and vials to prevent methanethiol (MeSH) oxidation, only hydrogen sulphide (H2S),
carbon disulphide (CS2), MeSH, and dimethyl sulphide (DMS) were the volatile sulfur
compounds developed during ripening. However, in figure (3) it can be seen that only small
and negligible amounts of dimethyl disulphide (DMDS) and dimethyl trisulphide (DMTS)
were rarely found in the chromatograms for the samples analyzed. Whereby, only the
development of MeSH, DMS and H2S will be discussed.
Because of the wide range of volatility and concentration of the VSC’s in the samples, and the
different selectivity of the CAR-PDMS SPME fiber, standard calibration curves had to be
calculated for each compound. It can be seen in figures 4 and 5 that good linear correlation
coefficients were obtained for H2S and DMS, but unfortunately not for MeSH. Whence, the
interpretation of results for this last compound was done based on the area ratio, between
MeSH and the internal standard EMS, instead of using its concentration during the ageing
process.
The results indicate that the heating treatment of cheese milk had a significant effect on the
VSC’s development during ripening. In both cases, pasteurization and heat-shock treatments,
temperature influenced the VSC’s concentration during ripening, suggesting that NSLAB
population and denaturation of serum proteins of milk may have contributed to their
formation.
In figure 6, hydrogen sulfide did not show a steady development for the heat-shock treatment
whereas it increased as cheese aged for the pasteurization treatment. Indeed, during the initial
stage of the ageing process the difference between the treatments was not significant, however
after 6 months it became evident, and the samples made with pasteurized cheese milk
displayed a higher concentration of H2S than those made with heat-shocked milk. It is also
possible to see that in spite of the unclear difference between treatments during the first 6-7
months the concentration of H2S increased over time, reaching a plateau after 9 months in the
case of pasteurized milk samples. Moreover, in the case of the heat-shock samples, it was
difficult to establish any trend during the ripening.
60
Since H2S sensory threshold is 10 ppb in water (Rychlik et al. 1998) and its concentration
along the ageing process varied from 12 to 50 ppm in the case of pasteurized samples, and
from 20 to 35 ppm for the heat-shocked samples; it was possible to confirm its role as key
contributor to the cheddar cheese aroma. The appreciable higher concentration for the
pasteurized samples can be attributed to higher denaturation and incorporation of βlactoglobulins to the casein micelles. Indeed, Cysteine, a sulfur containing amino acid present
in limited amounts in caseins, is the main precursor of H2S and it can be generated from the
sulfydryl groups once thermal breakdown of cysteine takes place (Fennema and Damodaran
1996). Thus, the results obtained suggest that the higher temperature used in the pasteurization
treatment promotes more coagulation of whey protein, increasing the availability of cysteine
and consequently the development of H2S during the cheese ageing. Indeed, Hutton and
Patton 1952; K. R. Christensen and Reineccius 1992 reported that the concentration of H2S in
milk increases linearly with heating temperature.
In the case of MeSH, because of the limited sulfur content of amino acids in caseins, it was
initially expected a higher concentration of this compound for the pasteurization treatment
during the cheese ripening due to the possible higher inclusion of whey protein into the cheese
curd. However, the results in figure 7 indicate a clear difference between treatments and
surprisingly higher amounts of MeSH for the samples made with heat-shocked cheese milk. In
addition, this trend suggests that the difference between treatments can arise from the
enzymatic activity of LAB and survival NSLAB coupled with the effect of the treatments on
the chemical structure of milk.
The higher development of MeSH for the heat shock treatment can be related to a bigger
population of indigenous bacteria as consequence of the lower temperature employed,
resulting in a possible increase of activity of the enzymes L-methionine γ-lyase (Tanaka,
Esaki, and Soda 1985) and/or cystathionine β-lyase and γ-lyase (Alting et al. 1995) during
ripening. As a matter of fact, some of the potent odorants in the cheddar cheese profile result
from leucine and methionine degradation, and it has been proposed that MeSH can be produce
from L-methionine via: 1) lysis of the C-S bond by L-methionine γ-lyase and/or cystathionine
β-lyase or γ-lyase), or 2) through a two-step pathway that involves the transamination of L-
61
methionine in the presence of a-keto glutarate to form a-keto-g-methyl thiobutyrate (KMTB)
(Yvon, et al 1997; Gao et al. 1998; Amarita et al. 2001), followed by its enzymatic breakdown
to form 3-methylthiopropionaldehyde and MeSH. In addition, when methanethiol is produce
from cysteine through the β-elimination reaction hydrogen sulfide is produced (Dobric et al.
2000; María Fernández et al. 2000)
Regarding the general absence in the results of this study of the sub-products DMDS and
DMTS resulting from the oxidation of MeSH, it might be possible to say based on the
thorough sample preparation work and the fact that cheese has a low redox potential, -150 to 200 mV, (Donald J. Manning and Moore 1979; Green and Manning 1982), that these
compounds are not original odorants produce during the ripening of Cheddar cheese, resulting
in “negligible” amounts.
On the other hand DMS had high concentrations. The results in figure 8 show a significant
difference between treatments, and it can be seen a steady increase of DMS for both type of
samples during the ripening stage, and higher amounts of DMS for the cheese prepared with
heat-shocked milk. The DMS concentration for most of the samples evaluated was higher than
its sensory threshold, 2ppm in water (Rychlik et al. 1998), demonstrating that this compound
is an important odorant in the Cheddar cheese matrix. The concentrations in pasteurized milk
varied from 1 to 14 ppm, while those for the heat-shock samples changed from 8 to 45 ppm,
which are amounts comparable to the ones obtained by Burbank and Qian (2008).
As far as we know there is no a proved and clear mechanism for the generation of DMS
during cheese ripening. However, it is well known that DMS concentration in raw milk is
significant and it is influenced by the diet of the cows (Manning et al. 1976; Forss 1979).
Furthermore, it is known that DMS can be generated from sulfhydryl group of milk proteins,
mainly β-lactoglobulin and if present
the milk fat globule membrane proteins, where
methionine is most likely the precursor for its generation (Datta et al. 2002). Therefore it is
expected that DMS is a natural component of the cheese milk, either raw or heated. Moreover,
lately in the work of (R. de Wit and Nieuwenhuijse 2008), it has been reported even the
possibility of a oxidation of MeSH into DMS and H2S (discussed later in the next chapter).
62
Figure 3 VSC's chromatogram Heat –schocked Vs Pasteurized milk
Figure 5 Calibration Curve H2S
Figure 7 Development of MeSH in Heat-schocked
and Pasteurized milk
Figure 4 Calibration curve DMS
Figure 6 Development of H2S in Heat-schocked
and Pasteurized milk
63
Figure 8 Development of DMS in Heat-schocked
and Pasteurized milk
Effect of treatment on Proteolysis
Cheddar cheese samples manufactured from pasteurized milk (72C for 15 sec) and heatshocked milk (66C for 30 sec) were examined during an 18 month ripening period by
measurements of Total Kjeldahl Nitrogen (TKN) of the WSN, TCA-SN and PTA-SN
fractions, RP-HPLC peptide profile of the WSN fraction and Urea-PAGE peptide profile of
the water insoluble fraction. The results revealed differences in the rate and pattern of
proteolysis. Indeed, from the three methods employed to evaluate the extend of proteolysis,
the TKN measurements and the peptide profile analysis by RP-HPLC of the WSN fraction,
were more effective than the Urea-PAGE analysis for discriminating the impact of the heat
treatment of cheese milk between samples. As a matter of fact, it was possible to observe
considerable differences in the primary and secondary proteolysis.
Soluble Nitrogen Fractions and TKN
Alteration to the ongoing proteolysis due to the heat treatment of cheese milk was obvious by
making use of the described fractionation scheme, which is based on the fact that extractability
of nitrogen compounds depend on pH (Ardö and Frederiksberg 1999; Sousa, Ardö, and
McSweeney 2001). Thus, it can be seen in figures 9 to 13, that the nitrogen level in the
recovery fractions increased during time and was different regarding the heating treatment of
cheese milk. Among the nitrogen indices proposed in this part of the study, the WSN fraction
includes all casein breakdown products, but native caseins and high molecular weight
peptides; the 12% trichloracetic TCA-SN fraction contains small peptides and FAA; and PTA-
64
SN fraction, which contains the smallest peptides (600 Da) and FAA (T. M. I. E. Christensen,
Bech, and Werner 1991).
In the figures 9 and 10, the levels of WSN were the highest for both types of cheeses, and had
values about 4 times higher than those of the TCA-SN fraction and 10 times higher in
comparison to those of the PTA-SN fraction, except for those points corresponding to the first
4 months of maturation, which were much higher, close to 6 and 12 times respectively, which
might be related to the primary proteolysis. As a matter of fact after 6 months of aging, the
values of TCA-SN and PTA-SN increased substantially for the all the samples. On the other
hand, it is possible to appreciate in figures 11, 12 and 13, how the values were significantly
affected by the treatment, and the last nitrogen contents in every fraction were 1.5 to 2 fold
higher than those at the beginning of the observation.
Regarding the levels of the WSN fraction, figure 11 shows how the values increased over
time, and although the amounts were similar at the beginning of the ageing process for both
types of samples, they were higher for those samples prepared with heat-shocked cheese milk.
In addition, the result in this figure also confirm the use of this fraction as an effective index of
primary proteolysis (Bansal, Piraino, and McSweeney 2009), and showed how the heating
treatment of the cheese milk, regardless the intensity and resemblance of it, can cause an
appreciable difference in the maturation of the cheese samples.This can be related to the fact
that the main enzymatic activity during primary proteolysis is provided by the remaining
rennet and the indigenous milk proteinase (S. Visser 1993; Paul L. H. McSweeney et al.
1994). Also, this is believed to do with the altered accessibility of the retained Chymosin to
the substrates, αs1 and κ caseins (Mulvihill and Fox 1979; P. F. Fox 1989), and with the impact
of the treatment on the Plasmin activity, which in spite of being stable at elevated temperature,
is definitively influenced by the heating protocol (Alichanidis, Wrathall, and Andrews 1986;
Metwalli, de Jongh, and van Boekel 1998). Indeed, it is possible to suggest that the difference
between treatments could be attributed to better accessibility of Chymosin to the bonds
Leu101-Lys102, Phe32-Gly33 and Leu109-Glu110 of the large peptides.
The levels of TCA-SN in Figure 12 also illustrate a proteolysis development affected by the
heating treatment of cheese milk. The increment in the soluble nitrogen in this fraction for the
65
experimental cheeses revealed a trend similar to that of the WSN fraction. The pasteurization
of milk decreased the concentration of nitrogen in the fraction with respect to the results of the
heat-shocked milk samples. However, it is possible to observe that the evolution of this
fraction for both types of samples was more regular and sustained throughout ripening,
resulting in final values that are approximately 3 times higher than those at the beginning of
the observation. Also, despite the results were slightly lower than those values found by
(Voigt et al. 2012) for the control sample, they are still comparable.
Based on the fact that TCA is a fraction rich in small peptides of low and medium
hydrophobicity (Kuchroo & Fox 1982), which contain peptides or traces of peptides mostly
derived from the N-terminal half of β-casein and from the N-terminal half of αs1-casein
(Tanoj K. Singh, Fox, and Healy 1995; T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and
Healy 1997; Manuela Fernández, Singh, and Fox 1998) resulting from the starter and NSLAB
enzymatic activity (O’Keefe et al. 1978). It might be possible to suggest that the difference
between treatments in this fraction can be attributed to intracellular and extracellular
peptidases and amino peptidase of the LAB and NSLAB. Indeed, it has been reported that in
spite of the destruction by pasteurization of pathogens, coliforms, psychotrophs and the
reduction of most of the LAB strain by 6 log (Burton 1986), it has been found that one strain
of Lactobacillus casei var. casei was reduced by only 3.5 log and that thermophilic LAB were
definitively more resistant to this treatment, in addition to other facts such as that in Swiss
cheese only 4 of 60 lactobacilli isolated were eliminated at 71C for 18 s (Bassett and Slatter
1953; Niven, Buettner, and Evans 1954), and that Propionibacterium freudenreichii, found in
raw milk (Baer and Ryba 1992), survived heating at 62.8 C for 30 min. Therefore, it is evident
that in the case of a milder treatment of the cheese milk such as Heat-shock, it can be expected
to find a more significant growth of NSLAB in Cheddar Cheese, resulting in a TCA fraction
with higher content of soluble nitrogen.
On the other hand, Lau, Barbano, and Rasmussen 1990; Lau, Barbano, and Rasmussen 1991
found slightly higher level of moisture in cheddar cheese made from pasteurized milk in
comparison to cheese made from raw milk, which indicates that syneresis could be affected by
the treatment and can impact on the proteolysis pattern because of the levels of bound water
that can result in less protein hydrolysis. It was found in the work of Whetstine et al. 2007,
66
that Cheddar cheese blocks corresponding to the inner side of a 291 Kg block had smaller
moisture content and had faster proteolysis in comparison to those of the outer side,
characterized by higher moisture content and less protein hydrolysis, which might be reflected
in the mobility of small peptides that contribute to the TCA-SN fraction. Which is another
reason why it can be expected that the heat-shock treatment, which is milder, shows slightly
higher levels of soluble nitrogen in TCA fraction.
The PTA-SN fraction is an index of secondary proteolysis because it is mainly constituted by
very small peptides (<15 kDa) and amino acids of approximately 600 Da (Aston and Dulley
1982). The results for this fraction showed in Fig 13, indicate a similar trend to the other
fractions, where the values increased over time and showed faster proteolysis for the heatshocked samples. Moreover, it is possible to appreciate that in comparison to the other
fractions, the increment of nitrogen content was more regular and sturdy, displaying final
values that are 4 times higher than those at the begin of the observation. Additionally, it can be
seen that the values for both treatments were really tight during the first 4 months, which
coincides with the period of time corresponding to the primary proteolysis. In a similar way to
the TCA fraction, these results could be explain by the low efficiency of the milder heating
treatments on NSLAB, elucidating the role of milk flora in the rate of proteolysis during the
ripening stage.
In general, it is possible to appreciate that as cheese ages the total amount of water soluble
peptides increases. In addition, it is feasible to affirm that the fractionation using water is
sufficient to extract the majority of water soluble peptides to establish a fair comparison
between treatments through the recovered nitrogen. Some possible explanations to the
appreciable difference in the proteolysis between treatments has to do with 1) The increase in
plasmin activity, regardless if it was due to denaturation of the inhibitors of the plasminogen
activators, or because of the inactivation of plasmin inhibitors (Baer et al. Collin et al. 1990;
Bastian & Brown 1996); 2) The denaturation of whey protein and subsequent alteration of the
rennetability of milk, affecting the moisture content the and LAB enzymatic activity; 3) And
as it has been mentioned above, a low elimination of NSLAB activity.
67
Figure 9 TKN fractions heat-schocked cheese milk
Figure 10 TKN fractions pasteurized cheese milk
Figure 11 WSN heat-shocked and pasteurized cheese milk
68
Figure 12 TCA heat-schocked and pasteurized cheese milk
Figure 13 PTA-SN heat-schocked and pasteurized cheese milk
Peptide analysis by RP-HPLC
The peptide analysis by RP-HPLC is an alternative index to express the degree of secondary
proteolysis. In addition, it can be used in authenticity studies and optimization process
(Upadhyay et al. 2004).
In this case, in order to evaluate the impact of the heating treatments of the cheese milk on the
peptide profiles in the chromatograms, it was used a principal component analysis (PCA),
which is essentially a multivariate analysis tool use in descriptive statistics, to estimate the
69
linear relationship between variables when their number is very large (Chatfield and Collins
1981). The raw data from chromatograms were processed based on Piraino, Parente, and
McSweeney 2004 work, where the complexity of the profiles containing more than 70 peaks,
is initially reduced by establishing a set of time intervals according the elution time; for which
the area of each one is expressed as percentage of total area of the chromatogram. Then the
variability due to treatments, biological factors (ripening, cheese making process, milk quality,
etc) and technical factors (sampling, extraction steps, and measurement of peak and intervals
area), is measured and analyzed through the variation of the contributing eigenvalues in the
correlation matrix for the principal components of the resulting model.
Two treatments were compared through a randomized block design with three replicates. The
peptide profiles revealed differences that were supported by the PCA results. Indeed, the PCA
analysis leads to a model with three principal components (PCs) that explain the 82.6% of the
variability of the data as can be seen in figure 14. However, after comparing in pairs the score
plots for any combination of the principal components, only the score plot for the PC1 Vs PC2
revealed a correlation. This is in agreement with the published results of (Benfeldt and
Sørensen 2001).
The first and second principal components explained the 73.1% of the variance. In the score
plot for this two PC’s (Fig. 15 and 16), it is possible to see how the PC1 differentiates the
samples according to their age, while PC2 represent mainly the contribution of the heating
treatment to the variation. Thus, the figures display higher scores for those samples with
longer ripening that correspond to the heat-shock treatment, which suggest that proteolysis is
slower for cheeses made from pasteurized milk.
The loading plot in figure 17, presents the projection of the eluting intervals on the PC1 and
PC2, and allows determining correlations between variables and their effect on the amount of
peptides eluting within certain retention times. Regarding the segments from 12-20 and 20-25
minutes, it can be seen that they have high scores for the PC1 and values close to 0 for PC2,
which suggest that the amount of peptides eluting in this zones increases during time and is
relatively unaffected by the type of heat treatment of the cheese milk. This interpretation could
be explain by the fact that the segments from 12-20 and 20-25 are mainly composed by
70
peptides products from the action of Chymosin on αs1-CN and k-caseins. In addition, there are
other peptides resulting from the action of cell envelope proteinase (CEP) from LAB such as
αs1-CN(f1-9) and αs1-CN(f1-13), which accumulate during ripening (Tanoj K. Singh, Fox,
and Healy 1997; Manuela Fernández, Singh, and Fox 1998). This indicates that the formation
of these hydrophilic peptides is mainly dependent on proteolytic systems that are not affected
by the heat treatment of the cheese milk, and it is probable that the subtle differences between
samples can be attributed to the amount of NSLA that survive after the treatment.
In contrast, the segments 25-30 and 30-35 got negative loadings for the PC1, meaning that the
relative amount of peptides eluting in this intervals decrease over time. This is believed to do
with the breakdown of the αs1-CN, αs2-CN, αs1-CN (f24-199) and β-CN peptides, which is
related to enzyme activity proportionate by the rennet and indigenous milk enzymes.
However, it is important to keep in mind that the concentration of αs2-casein appears to
decrease during ripening, but no large peptides derived from αs2-casein have been reported
yet (Mooney et al. 1998), and only a few small peptides have been identified in the WSE (T.K.
Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox
1998). Indeed, these observations are in agreement with previous works in Danbo cheese
(Benfeldt et al. 1997; Benfeldt and Sørensen 2001), and make sense since the breakdown of
caseins by chymosin and plasmin takes place along the ripening (especially during the primary
proteolysis).
Nonetheless, the PC2 loading scores for these 2 intervals reveal a mild reduction in the
amount of peptides eluting within these retention times as the temperature of the heating
treatment increases. Certainly, it is expected that the Plasminogen activation system and the
plasma-derived proteinase inhibitors respond to any thermal change. As a matter of fact, while
Plasmin and the complex Plasminogen activation system is related to the casein micelles and
are stable at high temperatures (S. Christensen et al. 1995), the proteinase inhibitory activity in
milk is associated to the serum phase (Heegaard, Rasmussen, and Andreasen 1994), which is
susceptible to heat denaturation due to its secondary and tertiary structures. Thus, it can be
thought that a lower treatment temperature causes less denaturation and association of serum
proteins to the casein micelles, and consequently less integration of inhibitors of Plasminogen
activators into the cheese. Additionally, less thermally induced thiol disulphide exchange
71
between Plasmin and β-lactoglobulin can be expected, (P. F. Fox and Stepaniak 1993),
resulting in higher Plasmin activity, reflected in the loading scores for these intervals that
contain the αs2-CN and β-CN peaks. Also, a possible reduced activity of Chymosin on the
αs1-CN might be expected and can be related to the association of β-lactoglobulin to the
casein micelle, which makes more difficult the accessibility of Chymosin to its active sites,
leading to a subtle difference between the treatments, which is appreciable as well in the score
for the 25-30 that is slightly off from zero regarding to the PC2.
On the other hand, the intervals 35-40 and 40-45 exhibit slightly positive loading scores in
relation to the PC1, and fairly negative ones regarding to the PC2. This suggests that the
relative amount of peptides in this elution fractions raise with the cheese ageing and
diminishes as the treatment temperature is increased. The reasons for this behavior can be
found in the fact that these fractions contain mainly hydrophobic peptides such as: 1) the
fragments β-CN(f29–209), β-CN(f106–209) and β-CN(f108–209) (γ1, γ 2, and γ 3,
respectively), whose concentrations increase during ripening (Farkye and Fox 1990) and are
the result from the hydrolysis of β-Casein by Plasmin at Lys28-Lys29, Lys105-Gln106 and
Lys107-Glu108 bonds; 2) the peptides αs1-CN(f93–?), αs1-CN(f24–30), αs1-CN(f26–32),
αs1-CN(f26–34) resulting from the hydrolysis of the peptide αs1-CN(f24–199) by Chymosin,
CEP and aminopeptidase (T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1995; Tanoj
K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox 1998); 3) peptides αs2CN(f204–207), which is a C-terminal residue and product of lactococcal CEP (T.K. Singh et
al. 1994). Furthermore, the slightly difference between heating treatments of cheese milk
regarding the amount of peptides eluting during these intervals can be related as well to a
higher or total inactivation of the acid proteinase Cathepsin D and the leucocytes proteinase by
pasteurization. Cathepsin D is more active on αs1-CN than on β-CN, and results in similar
breakdown products than Chymosin does, specially on the peptide αs1-CN(f24–199) (P. F.
Fox and McSweeney 1996; Larsen et al. 1996). This enzyme is completely inactivated at pH
7.0 and temperatures higher than 60C for 10 min, whereby it can be expected that the
pasteurization treatment had a higher impact in its final proteolytic activity during ripening
(Ducastaing et al.,1976). On the other hand, proteolytic activity of leucocyte proteinases is
similar to that of cathepsin D, and it may be a source of cathepsin D in milk by itself (P.L.H.
McSweeney, Fox, and Olson 1995).
72
Thus, because of the complete loss of leucocyte vitality by pasteurization (Grieve and Kitchen
1985), chemical alteration of the milk structure, and lower efficiency of the treatments in the
elimination of native milk enzymes (Lau, Barbano, and Rasmussen 1991), it could be
suggested that a “milder treatment” will be reflected on the scores for the loading plot and the
scores plot. However, it is important to keep in mind that only 50% of the of β-casein in
Cheddar cheese is hydrolyzed (T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1995;
Tanoj K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox 1998), and a still
unknown role of the alkaline phosphatase enzyme, which is inactivated at 70C for 15 sec, may
have a contribution to the slight difference between treatments (Griffiths 1986).
Regarding the scores for the intervals between 45-65 minutes, in spite they are composed by
hydrophobic compounds as well, they display a different trend to that shown in the intervals
between 35 to 45 min, which suggest that a considerable amount of amino acids might elute in
this zone, resulting from enzyme activity that is not severely affected by the heating treatment.
However, it is possible to see in the loading plot that their amount decreases as cheese aged,
which might has to do with the fact that as cheese ages more caseins, long, medium and small
peptides are broken into smaller pieces that may or may not be water soluble and finally can
undergo catabolic reactions (Lau, Barbano, and Rasmussen 1991). In addition, it has been
thought that the material eluting in this region could correspond to high molecular mass
molecules or molecules that contain aromatic amino acids, which are characterized for being
very hydrophobic. Lau, Barbano, and Rasmussen 1990, proposed that the products of peptides
containing aromatic amino acids can be highly hydrophobic and no longer water soluble,
which may indicate smaller amounts of hydrophobic peptides as cheese aged (an observation
that fits to the results obtain in this work). A complement to this last explanation is that after
Aston and Creamer, 1986, demonstrate that the water-soluble nitrogen fraction is an important
contributor to the nonvolatile flavor of cheese, Allan J. Cliffe, Marks, and Mulholland 1993,
reported that fractions isolated by gel filtration of the water soluble extract of well matured
Cheddar cheese, may vary from bitterness in the higher molecular mass components to savory
for those of lower molecular mass. Therefore, bitter fractions can be related to material that
was eluting late on the RP-HPLC column since they can have higher hydrophobic interactions.
In addition, it has been said that bitterness in cheese is associated to the presence aromatic
73
amino acids in free form or as part of a peptide. Furthermore, based on earlier works that
describe these segments as fractions mainly composed by amino acids, it can be expected that
once they are in its free form they become precursors of flavor compounds through catabolic
reactions, resulting in decreasing amounts of components for these zones of the
chromatogram.
Figure 14 Scree plot of heat-shocked and pasteurized cheese milk
Figure 15 Score plot of heat-shocked and pasteurized cheese milk (age)
74
Figure 16 Score plot of heat-shocked and pasteurized cheese milk (temperature)
Figure 17 Loading plot for heat-shocked and pasteurized cheese milk
75
Figure 18 Peptide profile pasteurized cheese (A and B) Vs Heat-shocked (C and D) milk, for 2 and 12 months
A
C
B
D
Electrophoresis
The results of the urea-PAGE of the pH4.6 insoluble nitrogen fraction of experimental
Cheddar cheese in Figs 19 and 20, clearly exhibit the progressive change of αS1-CN into the
peptides αS1-CN (f24-199), αS1-CN (f121-199), αS1-CN (f99-199), and the β-CN into peptides
β-CN (f29-202), β-CN (f108-209) and β-CN (f106-209) during ripening. In addition bands of
γ-CNs became more noticeable after 4 months. Also it can be seen that apparently the
development of peptides from αS1-CN is faster than those from β-CN, which can be related to
primary proteolysis and the actual amount of β-CN that is hydrolyzed (only the 50%).
However, the representative urea-PAGE gel do not show any appreciable difference between
treatment, and barely minor differences are perceived between treatments after 14 months in
comparison to those of the early stages of ripening.
76
Figure 19 Urea PAGE for ripening of cheese made from heat-shocked milk
β -CN(f106-209)
β -CN(f29-209)
β -CN(f108-209)
β -CN
αs1-CN
αs1CN(f121-199)
αs1 -CN(f24-199)
2M
4M
8M
10 M
12 M
14 M
Figure 20 Urea PAGE for heat-schocked Vs pasteurized cheeses (12 months)
Heat-shock
Pasteurized
77
CONCLUSION
The present study demonstrates that the use of FFA profile, VSC’s profile, measurement of
the levels of the Total Kjeldahl nitrogen for the WSN, TCA-SN and PTA-SN fractions, and
the analysis of the RP-HPLC peptide profile of the WSN fraction by using a PCA, are
effective tools and indices of ripening to differentiate Cheddar cheese samples regarding to
age and type of heat treatment of the cheese milk. In the case of Urea-PAGE, it was
demonstrated that its use as index of primary proteolysis can be effective to differentiate
samples by their age; nonetheless it is clear that it is a method not adecuate to detect
differences between samples made from heat-shocked or pasteurized cheese milk unless it is
coupled to other systems to obtain electrophotograms. In addition, it was demonstrated that
proteolysis is faster for cheeses made with heat-shocked cheese milk since the results of
nitrogen level for all the 3 fractions analyzed were higher than those found for the pasteurized
cheese milk samples. This was supported by the PCA model obtained, which suggest
alterations to the structure of milk and to the amount of remaining indigenous bacteria. It was
found slightly higher levels of short and medium FFA; however, the difference FFA is not
significant. The amounts found for DMS, H2S and MeSH showed appreciable differences
between samples, and it can be seen that cheeses made from heat-shocked milk undergo faster
catabolism of sulfur containing amino acids such as methionine and cysteine. Another
interesting observation is that it seems like the DMDS and DMTS, reported in previos works
as important contributors to cheese flavor, can be artifacts from extraction and separation
procedures, consecuence of the oxidation of MeSH, rather than metabolites from the ripening
of Cheddar cheese.
REFERENCES
Alichanidis, Efstathios, Julia H. M. Wrathall, and Anthony T. Andrews. 1986. “Heat Stability
of Plasmin (milk Proteinase) and Plasminogen.” Journal of Dairy Research V53 (2):
pp. 259–269.
Alting, A. C., W. Engels, S. van Schalkwijk, and F. A. Exterkate. 1995. “Purification and
Characterization of Cystathionine (beta)-Lyase from Lactococcus Lactis Subsp.
Cremoris B78 and Its Possible Role in Flavor Development in Cheese.” Applied and
Environmental Microbiology V61 (11) (November 1): pp. 4037–4042.
Amarita, F., T. Requena, G. Taborda, L. Amigo, and C. Pelaez. 2001. “Lactobacillus Casei
and Lactobacillus Plantarum Initiate Catabolism of Methionine by Transamination.”
Journal of Applied Microbiology V90 (6): pp. 971–978.
78
Andersen, Lene T., Ylva Ardö, and Wender L.P. Bredie. 2010. “Study of Taste-active
Compounds in the Water-soluble Extract of Mature Cheddar Cheese.” International
Dairy Journal V20 (8) (August): pp. 528–536.
Anderson, D.L., V.V. Mistry, R.L. Brandsma, and K.A. Baldwin. 1993. “Reduced Fat
Cheddar Cheese from Condensed Milk. 1. Manufacture, Composition, and Ripening.”
Journal of Dairy Science V76 (10) (October): pp. 2832–2844.
Ardö, Y., and Frederiksberg. 1999. “Evaluating Proteolysis by Analysing the N Content of
Cheese Fractions.” Bulletin - International Dairy Federation V337: pp. 4–9.
Arora, Gulshan, Francois Cormier, and Byong Lee. 1995. “Analysis of Odor-Active Volatiles
in Cheddar Cheese Headspace by Multidimensional GC/MS/Sniffing.” J. Agric. Food
Chem. V43 (3): pp. 748–752.
Aston, J. W., and L. K. Creamer. 1986. “Contribution of the components of the water-soluble
fraction to the flavour of Cheddar cheese.” New Zealand Journal of Dairy Science and
Technology V21 (3): pp. 229–248.
Aston, J. W., and John R. Dulley. 1982. “Cheddar cheese flavor review.” Australian Journal
of Dairy Technology V37 (2) (June): pp. 59–64.
Avsar, Y.K., Y. Karagul-Yuceer, M.A. Drake, T.K. Singh, Y. Yoon, and K.R. Cadwallader.
2004. “Characterization of Nutty Flavor in Cheddar Cheese.” Journal of Dairy
Science V87 (7) (July): pp. 1999–2010.
Ayad, Eman H.E, Annette Verheul, Catrienus de Jong, Jan T.M Wouters, and Gerrit Smit.
1999. “Flavour Forming Abilities and Amino Acid Requirements of Lactococcus
Lactis Strains Isolated from Artisanal and Non-dairy Origin.” International Dairy
Journal V9 (10) (October): pp. 725–735.
Baer, A., and I. Ryba. 1992. “Serological Identification of Propionibacteria in Milk and
Cheese Samples.” International Dairy Journal V2 (5): pp. 299–310.
Balcão, Victor M., and F.Xavier Malcata. 1998. “Lipase Catalyzed Modification of Milkfat.”
Biotechnology Advances V16 (2) (March): pp. 309–341.
Bansal, N, Paolo Piraino, and P. L. H. McSweeney. 2009. “Determination of Proteolysis in
Cheese.” In Handbook of Dairy Foods Analysis, pp. 405–417. CRC Press.
Bassett, Harold J., and W.L. Slatter. 1953. “The Heat Resistance of Some Lactobacilli Found
in Swiss Cheese.” Journal of Dairy Science V36 (4) (April): pp. 409–413.
Bastian, E.D., C.G. Lo, and K.M.M. David. 1997. “Plasminogen Activation in Cheese Milk:
Influence on Swiss Cheese Ripening.” Journal of Dairy Science 80 (2) (February):
245–251.
Bastian, Eric D., and Rodney J. Brown. 1996. “Plasmin in Milk and Dairy Products: An
Update.” International Dairy Journal V6 (5) (May): pp. 435–457.
Benfeldt, Connie, and John Sørensen. 2001. “Heat Treatment of Cheese Milk: Effect on
Proteolysis During Cheese Ripening.” International Dairy Journal V11 (4–7) (July
11): pp. 567–574.
Benfeldt, Connie, John Sørensen, Katrine H. Ellegård, and Torben E. Petersen. 1997. “Heat
Treatment of Cheese Milk: Effect on Plasmin Activity and Proteolysis During Cheese
Ripening.” International Dairy Journal 7 (11) (October 24): 723–731.
Benkler, K. F., and G. A. Reineccius. 1980. “FLAVOR ISOLATION FROM FATTY FOODS
VIA SOLVENT EXTRACTION AND MEMBRANE DIALYSIS.” Journal of Food
Science V45 (4): pp. 1084–1085.
Bentley, Ronald, and Thomas G. Chasteen. 2004. “Environmental VOSCs––formation and
Degradation of Dimethyl Sulfide, Methanethiol and Related Materials.” Chemosphere
V55 (3) (April): pp. 291–317.
79
Beresford, Tom P, Nora A Fitzsimons, Noelle L Brennan, and Tim M Cogan. 2001. “Recent
Advances in Cheese Microbiology.” International Dairy Journal V11 (4–7) (July 11):
pp. 259–274.
Beuvier, E., and S. Buchin. 2004. “Raw Milk Cheeses.” In Cheese Chemistry, Physics and
Microbiology, Volume 1:319–345. Academic Press.
Bills, D.D., and E.A. Day. 1964. “Determination of the Major Free Fatty Acids of Cheddar
Cheese.” Journal of Dairy Science V47 (7) (July): pp. 733–738.
Burbank, Helen M., and Michael C. Qian. 2005. “Volatile Sulfur Compounds in Cheddar
Cheese Determined by Headspace Solid-phase Microextraction and Gas
Chromatograph-pulsed Flame Photometric Detection.” Journal of Chromatography A
V1066 (1–2) (February 25): pp. 149–157.
Burbank, Helen, and Michael C. Qian. 2008. “Development of Volatile Sulfur Compounds in
Heat-shocked and Pasteurized Milk Cheese.” International Dairy Journal V18 (8)
(August): pp. 811–818.
Burton, H. 1986. “Microbiological aspects.” (200): 9–14.
Careri, M, and A Mangia. 2003. “Analysis of Food Proteins and Peptides by Chromatography
and Mass Spectrometry.” Journal of Chromatography A V1000 (1-2) (June 6): pp.
609–635.
Carunchia Whetstine, M E, P J Luck, M A Drake, E A Foegeding, P D Gerard, and D M
Barbano. 2007. “Characterization of Flavor and Texture Development Within Large
(291 Kg) Blocks of Cheddar Cheese.” Journal of Dairy Science V90 (7) (July): pp.
3091–3109.
Carunchia Whetstine, Mary E., Keith R. Cadwallader, and MaryAnne Drake. 2005.
“Characterization of Aroma Compounds Responsible for the Rosy/Floral Flavor in
Cheddar Cheese.” J. Agric. Food Chem. V53 (8): pp. 3126–3132.
del Castillo-Lozano, M. López, S. Mansour, R. Tâche, P. Bonnarme, and S. Landaud. 2008.
“The Effect of Cysteine on Production of Volatile Sulphur Compounds by Cheeseripening Bacteria.” International Journal of Food Microbiology V 122 (3) (March
20): pp. 321–327.
Chaintreau, Alain. 2001. “Simultaneous Distillation–extraction: From Birth to Maturity—
review.” Flavour and Fragrance Journal V16 (2): pp. 136–148.
Chatfield, Chris, and A. Collins. 1981. “Finding New Underlying Variables.” In Introduction
to Multivariate Analysis, pp. 57–78. CRC Press.
Chavarri, Felisa, Mailo Virto, Celia Martin, Ana I. Nájera, Arantza Santisteban, Luis J. R.
Barrón, and Mertxe De Renobales. 1997. “Determination of Free Fatty Acids in
Cheese: Comparison of Two Analytical Methods.” Journal of Dairy Research V64
(3): pp. 445–452.
Chich, Jean-François, Karine Marchesseau, and Jean-Claude Gripon. 1997. “Intracellular
Esterase from Lactococcus Lactis Subsp. Lactis NCDO 763: Purification and
Characterization.” International Dairy Journal V7 (2-3) (February): pp. 169–174.
Chopin, Alain. 1993. “Organization and Regulation of Genes for Amino Acid Biosynthesis in
Lactic Acid Bacteria.” FEMS Microbiology Reviews V12 (1–3) (September): pp. 21–
37.
Christensen J.E., Johnson M.E., and Steele J.L. 1995. “Production of Cheddar Cheese Using a
Lactococcus Lactis Ssp. Cremoris SK11 Derivative with Enhanced Aminopeptidase
Activity.” International Dairy Journal V5 (4): pp. 367–379.
80
Christensen, Jeffrey E., Edward G. Dudley, Jeffrey A. Pederson, and James L. Steele. 1999.
“Peptidases and Amino Acid Catabolism in Lactic Acid Bacteria.” Antonie Van
Leeuwenhoek V76 (1) (1): pp. 217–246.
Christensen, K.R., and G.A. Reineccius. 1992. “Gas Chromatographic Analysis of Volatile
Sulfur Compounds from Heated Milk Using Static Headspace Sampling.” Journal of
Dairy Science V75 (8) (August): pp. 2098–2104.
Christensen, Søren, Tove Wiegers, John Hermansen, and Lars Sottrup-Jensen. 1995. “Plasmaderived Protease Inhibitors in Bovine Milk.” International Dairy Journal V5 (5): pp.
439–449.
Christensen, T. M. I. E., A.-M. Bech, and H. Werner. 1991. “Methods for crude fractionation
(extraction and precipitation) of nitrogen components in cheese.” Bulletin International Dairy Federation (261): pp. 4–9.
Christensen, Tove M. I. E., Kristian R. Kristiansen, and Jesper S. Madsen. 1989. “Proteolysis
in Cheese Investigated by High Performance Liquid Chromatography.” Journal of
Dairy Research V56 (5): pp. 823–828.
Cliffe, A.J., D. Revell, and B.A. Law. 1989. “A Method for the Reverse Phase High
Performance Liquid Chromatography of Peptides from Cheddar Cheese.” Food
Chemistry 34 (2): 147–160.
Cliffe, Allan J., Jonathan D. Marks, and Francis Mulholland. 1993. “Isolation and
Characterization of Non-volatile Flavours from Cheese: Peptide Profile of Flavour
Fractions from Cheddar Cheese, Determined by Reverse-phase High-performance
Liquid Chromatography.” International Dairy Journal V3 (4–6): pp 379–387.
Collin, J. C., D. le Bars, I. Ryba, and A. Bear. 1990. “Comparative measurements of the
plasminogen-plasmin system of milk.” International Dairy Federation.
Collins, Y.F., P.L.H. McSweeney, and M.G. Wilkinson. 2004. “Lipolysis and Catabolism of
Fatty Acids in Cheese.” In Cheese Chemistry, Physics and Microbiology, Volume
1:373–389. Academic Press.
Collins, Yvonne F., Paul L.H. McSweeney, and Martin G. Wilkinson. 2003 “Lipolysis and
Free Fatty Acid Catabolism in Cheese: a Review of Current Knowledge.”
International Dairy Journal V13 (11): pp. 841–866.
Considine, T., A. Healy, A.L. Kelly, and P.L.H. McSweeney. 1999. “Proteolytic Specificity of
Elastase on Bovine Β-casein.” Food Chemistry V66 (4) (September): pp. 463–470.
Considine, T., Á. Healy, A.L. Kelly, and P.L.H. McSweeney. 2000. “Proteolytic Specificity of
Elastase on Bovine Αs1-casein.” Food Chemistry V69 (1) (April): pp. 19–26.
doi:10.1016/S0308-8146(99)00217-4.
Considine, T., Á. Healy, A.L. Kelly, and P.L.H. McSweeney. 2004. “Hydrolysis of Bovine
Caseins by Cathepsin B, a Cysteine Proteinase Indigenous to Milk.” International
Dairy Journal V14 (2) (February): pp. 117–124.
Crow, V.L., T. Coolbear, P.K. Gopal, F.G. Martley, L.L. McKay, and H. Riepe. 1995. “The
Role of Autolysis of Lactic Acid Bacteria in the Ripening of Cheese.” International
Dairy Journal V5 (8): pp. 855–875.
Crow, Vaughan, Brian Curry, and Michelle Hayes. 2001. “The Ecology of Non-starter Lactic
Acid Bacteria (NSLAB) and Their Use as Adjuncts in New Zealand Cheddar.”
International Dairy Journal V11 (4-7) (4–7) (July 11): pp 275–283.
Cserháti, Tibor. 2002. “Mass Spectrometric Detection in Chromatography. Trends and
Perspectives.” Biomedical Chromatography V16 (5): pp. 303–310.
81
Curioni, P.M.G., and J.O. Bosset. 2002. “Key Odorants in Various Cheese Types as
Determined by Gas Chromatography-olfactometry.” International Dairy Journal v12
(12): pp. 959–984.
Datta, N., A. J. Elliott, M. L. Perkins, and H. C. Deeth. 2002. “Ultra-high-temperature (UHT)
Treatment of Milk: Comparison of Direct and Indirect Modes of Heating” V57 (3):
pp. 211–227.
Deeth, H. C, and V. Touch. 2000a. “Methods for detecting lipase activity in milk and milk
products.” Australian Journal of Dairy Technology V55 (3): pp. 153–168.
Deeth, H. C., and V. Touch. 2000b. “Methods for detecting lipase activity in milk and milk
products.” Australian Journal of Dairy Technology V55 (3): pp. 153–168.
Dias, Benjamin, and Bart Weimer. 1998. “Conversion of Methionine to Thiols by Lactococci,
Lactobacilli, and Brevibacteria.” Applied and Environmental Microbiology V64 (9)
(September 1): pp. 3320–3326.
Dimos, Tony. 1992. “A comparative study by GC/MS of the flavour volatiles produced during
the maturation of full-fat and low-fat Cheddar cheese.”
Dirinck, P, and A De Winne. 1999. “Flavour Characterisation and Classification of Cheeses
by Gas Chromatographic–mass Spectrometric Profiling.” Journal of Chromatography
A V847 (1-2) (June 25): pp. 203–208.
Dobric, Nada, Gäetan K.Y. Limsowtin, Alan J. Hillier, Nicholas P.B. Dudman, and Barrie E.
Davidson. 2000. “Identification and Characterization of a Cystathionine Β/γ-lyase
from Lactococcus Lactis Ssp. Cremoris MG13631.” FEMS Microbiology Letters
V182 (2): pp. 249–254.
Drake, M.A., T.D. Boylston, K.D. Spence, and B.G. Swanson. 1996. “Chemical and Sensory
Effects of a Lactobacillus Adjunct in Cheddar Cheese.” Food Research International
V29 (3–4) (April): pp. 381–387.
El Soda, M., S.A. Madkor, and P.S. Tong. 2000. “Adjunct Cultures: Recent Developments
and Potential Significance to the Cheese Industry.” Journal of Dairy Science V83 (4)
(April): pp. 609–619.
El Soda, Morsi A. 1993. “The Role of Lactic Acid Bacteria in Accelerated Cheese Ripening.”
FEMS Microbiology Reviews V12 (1–3) (September): pp. 239–251.
El-Soda, M., S. A. Madkor, and P. S. Tong. 2000. “Evaluation of commercial adjuncts for use
in cheese ripening: 4. Comparison between attenuated and not attenuated lactobacilli.”
Milchwissenschaft V55 (5): pp. 260–263.
Engel, Wolfgang, Wolfgang Bahr, and P. Schieberle. 1999. “Solvent Assisted Flavour
Evaporation – a New and Versatile Technique for the Careful and Direct Isolation of
Aroma Compounds from Complex Food Matrices.” European Food Research and
Technology V209 (3): pp. 237–241.
Engels, W.J.M., R. Dekker, C. de Jong, R. Neeter, and S. Visser. 1997. “A Comparative Study
of Volatile Compounds in the Water-soluble Fraction of Various Types of Ripened
Cheese.” International Dairy Journal V7 (4) (4) (April): pp. 255–263.
Exterkate, Fred A.., Arno C.. Alting, and Charles J. Slangen. 1995. “Conversion of αs1Casein-(24-199)-Fragment and β-Casein Under Cheese Conditions by Chymosin and
Starter Peptidases.” Systematic and Applied Microbiology V18 (1) (1): pp. 7–12.
Fang, Yu, and Michael C. Qian. 2005. “Sensitive Quantification of Sulfur Compounds in
Wine by Headspace Solid-phase Microextraction Technique.” Journal of
Chromatography A V1080 (2) (July 8): pp. 177–185.
Farkye, Nana Y., and Patrick F. Fox. 1990. “Observations on Plasmin Activity in Cheese.”
Journal of Dairy Research V57 (03): pp. 413–418.
82
FDA. 2011. “Code of Federal Regulations Title 21.”
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr.
Fennema, Owen R., and Damodaran. 1996. “Amino Acids, Peptides, and Proteins.” In Food
Chemistry, pp. 322–425. CRC Press.
Fernández, Manuela, Tanoj K. Singh, and Patrick F. Fox. 1998. “Isolation and Identification
of Peptides from the Diafiltration Permeate of the Water-Soluble Fraction of Cheddar
Cheese.” Journal of Agricultural and Food Chemistry V46 (11) (November 1): pp.
4512–4517.
Fernández, María, Wim van Doesburg, Ger A. M. Rutten, Joey D. Marugg, Arno C. Alting,
Richard van Kranenburg, and Oscar P. Kuipers. 2000. “Molecular and Functional
Analyses of themetC Gene of Lactococcus Lactis, Encoding Cystathionine β-Lyase.”
Applied and Environmental Microbiology V66 (1) (January 1): pp. 42–48.
Fernández-García, Estrella. 1996. “Use of Headspace Sampling in the Quantitative Analysis
of Artisanal Spanish Cheese Aroma.” Journal of Agricultural and Food Chemistry
V44 (7) (January 1): pp. 1833–1839.
Ferreira, A.D., and B.C. Viljoen. 2003. “Yeasts as Adjunct Starters in Matured Cheddar
Cheese.” International Journal of Food Microbiology V86 (1-2) (1–2) (September 1):
pp. 131–140.
Forde, Amanda, and Gerald F Fitzgerald. 2000. “Biotechnological Approaches to the
Understanding and Improvement of Mature Cheese Flavour.” Current Opinion in
Biotechnology V11 (5) (October 1): pp. 484–489.
Fox, P. F., T.P. Guinee, T. M. Cogan, and P. L. H. McSweeney. 2000. “Enzymatic
Coagulation of Milk.” In Fundamentals of Cheese Science, pp. 98135. Springer.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Fox, P. F., J. A. Lucey, and T. M. Cogan. 1990. “Glycolysis and Related Reactions During
Cheese Manufacture and Ripening.” Critical Reviews in Food Science and Nutrition
V29 (4) (January): pp. 237–253.
Fox, P. F., and P. L. H. McSweeney. 1996. “Proteolysis in Cheese During Ripening.” Food
Reviews International V12 (4): pp. 457–509.
Fox, P. F., and P. L. H. McSweeney. 1998. “Milk Lipids.” In Dairy Chemistry and
Biochemistry, pp. 67–141. Springer.
Fox, P. F., J. M. Wallace, S. Morgan, C. M. Lynch, E. J. Niland, and J. Tobin. 1996.
“Acceleration of Cheese Ripening.” Antonie Van Leeuwenhoek 70 (2): 271–297.
Fox, P.F. 1989. “Proteolysis During Cheese Manufacture and Ripening.” Journal of Dairy
Science V72 (6) (6) (June): pp. 1379–1400.
Fox, P.F., and P.L.H. McSweeney. 2004. “Cheese: An Overview.” In Cheese Chemistry,
Physics and Microbiology, Volume 1:1–18. Academic Press.
Fox, P.F., and L. Stepaniak. 1993. “Enzymes in Cheese Technology.” International Dairy
Journal V3 (4–6): pp. 509–530.
Fox, PF, PLH McSweeney, and CM Lynch. 1998. “Significance of Non-starter Lactic Acid
Bacteria in Cheddar Cheese.” Australian Journal of Dairy Technology V53 (2): pp. 7–
8.
Gao, Song, Ed S. Mooberry, and James L. Steele. 1998. “Use of 13C Nuclear Magnetic
Resonance and Gas Chromatography To Examine Methionine Catabolism by
Lactococci.” Applied and Environmental Microbiology V64 (12) (December 1): pp.
4670–4675.
83
Gomez, Maria J., Sonia Garde, Pilar Gaya, Margarita Medina, and Manuel Nuñez. 1997.
“Relationship Between Level of Hydrophobic Peptides and Bitterness in Cheese Made
from Pasteurized and Raw Milk.” Journal of Dairy Research V64 (2): pp. 289–297.
Grandison, Alistair S., and Graeme D. Ford. 1986. “Effects of Variations in Somatic Cell
Count on the Rennet Coagulation Properties of Milk and on the Yield, Composition
and Quality of Cheddar Cheese.” Journal of Dairy Research V53 (4): pp. 645–655.
Grappin, Rémy, and Eric Beuvier. 1997. “Possible Implications of Milk Pasteurization on the
Manufacture and Sensory Quality of Ripened Cheese.” International Dairy Journal
V7 (12) (November 21): pp. 751–761.
Green, Margaret L., and Donald J. Manning. 1982. “Development of Texture and Flavour in
Cheese and Other Fermented Products.” Journal of Dairy Research V49 (04): pp.
737–748.
Grieve, Paul A., and Barry J. Kitchen. 1985. “Proteolysis in Milk: The Significance of
Proteinases Originating from Milk Leucocytes and a Comparison of the Action of
Leucocyte, Bacterial and Natural Milk Proteinases on Casein.” Journal of Dairy
Research 52 (1): 101–112.
Griffiths, M. W. 1986. “Use of milk enzymes as indices of heat treatment.” Journal of Food
Protection V49 (9): pp. 696–705.
Gripon, Jean-Claude, V Monnet, G Lambert, and M. J. Desmazeaud. 1991. “Microbial
enzymes in cheese ripening.” In Food enzymology, vol 1.
Guinee, T P. 2004. “Salting and the Role of Salt in Cheese.” International Journal of Dairy
Technology V57 (2-3): pp. 99–109.
Habibi‐Najafi, Mohammad B., Byong H. Lee, and Barry Law. 1996. “Bitterness in Cheese: A
Review.” Critical Reviews in Food Science and Nutrition V36 (5): pp. 397–411.
Heegaard, Christian W., Lone K. Rasmussen, and Peter A. Andreasen. 1994. “The
Plasminogen Activation System in Bovine Milk: Differential Localization of Tissuetype Plasminogen Activator and Urokinase in Milk Fractions Is Caused by Binding to
Casein and Urokinase Receptor.” Biochimica Et Biophysica Acta (BBA) - Molecular
Cell Research V1222 (1) (May 26): pp. 45–55.
Hickey, D.K., K.N. Kilcawley, T.P. Beresford, and M.G. Wilkinson. 2007. “Lipolysis in
Cheddar Cheese Made from Raw, Thermized, and Pasteurized Milks.” Journal of
Dairy Science V90 (1) (January): pp. 47–56.
Holland, Ross, and Tim Coolbear. 1996. “Purification of Tributyrin Esterase from
Lactococcus Lactis Subsp. Cremoris E8.” Journal of Dairy Research V63 (1): pp.
131–140.
Home, D.S., and J.M. Banks. 2004. “Rennet-induced Coagulation of Milk.” In Cheese
Chemistry, Physics and Microbiology, Volume 1:pp. 47–70. Academic Press.
Huffman, L. M.;, and Kristoffersen. 1984. “Role of lactose in Cheddar cheese manufacturing
and ripening.” New Zealand Journal of Dairy Science and Technology V19 (2): pp.
151–162.
Hurley, M.J, B.M O’Driscoll, A.L Kelly, and P.L.H McSweeney. 1999. “Novel Assay for the
Determination of Residual Coagulant Activity in Cheese.” International Dairy
Journal V9 (8) (August): pp 553–558.
Hutton, J.T., and S. Patton. 1952. “The Origin of Sulfhydryl Groups in Milk Proteins and
Their Contributions to ‘Cooked’ Flavor.” Journal of Dairy Science V35 (8) (August):
pp. 699–705
84
Jolly, Ramesh C., and Frank V. Kosikowski. 1975. “Quantification of Lactones in Ripening
Pasteurized Milk Blue Cheese Containing Added Microbial Lipases.” J. Agric. Food
Chem. V23 (6): pp. 1175–1176.
De Jong, Catrienus, and Herman T. Badings. 1990. “Determination of Free Fatty Acids in
Milk and Cheese Procedures for Extraction, Clean up, and Capillary Gas
Chromatographic Analysis.” Journal of High Resolution Chromatography V13 (2):
pp. 94–98.
K. Arfi, H.Spinnler, R. Tache, and P. bonnarme. 2002. “Production of Volatile Compounds by
Cheese-ripening Yeasts: Requirement for a Methanethiol Donor for S-methyl
Thioacetate Synthesis by Kluyveromyces Lactis.” Applied Microbiology and
Biotechnology V58 (4): pp. 503–510.
Knecht, E. 1999. “Lysomomal Cysteine Proteinases (2nd Edition).” Cell Biology International
V23 (8) (August): pp 590–591.
Laan, Harry, Saw Eng Tan, Paul Bruinenberg, Gaëtan Limsowtin, and Malcolm Broome.
1998. “Aminopeptidase Activities of Starter and Non-Starter Lactic Acid Bacteria
Under Simulated Cheddar Cheese Ripening Conditions.” International Dairy Journal
V8 (4) (April): pp. 267–274.
Lane, C.N., and P.F. Fox. 1996. “Contribution of Starter and Adjunct Lactobacilli to
Proteolysis in Cheddar Cheese During Ripening.” International Dairy Journal V6 (7)
(July): pp. 715–728.
Lane, Cáit N., and Patrick F. Fox. 1997. “Role of Starter Enzymes During Ripening of
Cheddar Cheese Made from Pasteurized Milk Under Controlled Microbiological
Conditions.” International Dairy Journal V7 (1) (January): pp. 55–63.
Larsen, Lotte B., Connie Benfeldt, Lone K. Rasmussen, and Torben E. Petersen. 1996.
“Bovine Milk Procathepsin D and Cathepsin D: Coagulation and Milk Protein
Degradation.” Journal of Dairy Research V63 (1): pp 119–130.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen. 1990. “Influence of
Pasteurization on Fat and Nitrogen Recoveries and Cheddar Cheese Yield.” Journal of
Dairy Science V73 (3) (March): pp. 561–570.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen.1991. “Influence of
Pasteurization of Milk on Protein Breakdown in Cheddar Cheese During Aging.”
Journal of Dairy Science V74 (3) (March): pp. 727–740.
Law, A. J. R., J. M. Banks, D. S. Horne, J. Leaver, and I. G. West. 1994. “Denaturation of the
whey proteins in heated milk and their incorporation into Cheddar cheese.”
Milchwissenschaft V49 (2): pp. 63–67.
Lee, Won-Jae, Dattatreya S. Banavara, Joanne E. Hughes, Jason K. Christiansen, James L.
Steele, Jeffery R. Broadbent, and Scott A. Rankin. 2007. “Role of Cystathionine βLyase in Catabolism of Amino Acids to Sulfur Volatiles by Genetic Variants of
Lactobacillus Helveticus CNRZ 32.” Applied and Environmental Microbiology V73
(9) (May 1): pp. 3034–3039.
Lin, J. C. C., and I. J. Jeon. 1985. “Headspace Gas Sampling/GC Method for the Quantitative
Analysis of Volatile Compounds in Cheese.” Journal of Food Science V50 (3): pp.
843–844.
Liu, Shao-Quan, Ross Holland, and Vaughan L Crow. 2001. “Purification and Properties of
Intracellular Esterases from Streptococcus Thermophilus.” International Dairy
Journal V11 (1–2) (January): pp. 27–35.
85
Lloyd, G. T., J. F. Horwood, and I. Barlow. 1980. “The effect of yoghurt culture YB on the
flavour and maturation of Cheddar cheese.” Australian Journal of Dairy Technology
V35 (4): pp. 137–139.
Lynch, C.M., P.L.H. McSweeney, P.F. Fox, T.M. Cogan, and F.D. Drinan. 1996.
“Manufacture of Cheddar Cheese with and Without Adjunct Lactobacilli Under
Controlled Microbiological Conditions.” International Dairy Journal V6 (8–9)
(August): pp. 851–867.
Lynch, C.M., D.D. Muir, J.M. Banks, P.L.H. McSweeney, and P.F. Fox. 1999. “Influence of
Adjunct Cultures of Lactobacillus Paracasei Ssp. Paracasei or Lactobacillus Plantarum
on Cheddar Cheese Ripening.” Journal of Dairy Science V82 (8) (August): pp. 1618–
1628.
Lynch, Joanna M., David M. Barbano, and J. Richard Fleming. 2002. “Determination of the
Total Nitrogen Content of Hard, Semihard, and Processed Cheese by the Kjeldahl
Method: Collaborative Study.” Journal of AOAC International V85 (2): pp. 445–455.
Madkor, S.A., P.S. Tong, and M. El Soda. 2000. “Ripening of Cheddar Cheese with Added
Attenuated Adjunct Cultures of Lactobacilli.” Journal of Dairy Science V83 (8)
(August): pp. 1684–1691.
Manning, D. J., Helen R. Chapman, and Zena D. Hosking. 1976. “The Production of Sulphur
Compounds in Cheddar Cheese and Their Significance in Flavour Development.”
Journal of Dairy Research V43 (02): pp. 313–320.
Manning, Donald J., and Carolyn Moore. 1979. “Headspace Analysis of Hard Cheeses.”
Journal of Dairy Research V46 (03): pp. 539–545.
McSweeney, P.L.H., P.F. Fox, J.A. Lucey, K.N. Jordan, and T.M. Cogan. 1993. “Contribution
of the Indigenous Microflora to the Maturation of Cheddar Cheese.” International
Dairy Journal V3 (7): pp. 613–634.
McSweeney, P.L.H., P.F. Fox, and N.F. Olson. 1995. “Proteolysis of Bovine Caseins by
Cathepsin D: Preliminary Observations and Comparison with Chymosin.”
International Dairy Journal V5 (4): pp. 321–336.
McSweeney, Paul L. H., Sylvie Pochet, Patrick F. Fox, and Aine Healy. 1994. “Partial
Identification of Peptides from the Water-Insoluble Fraction of Cheddar Cheese.”
Journal of Dairy Research V61 (4): pp. 587–590.
McSweeney, Paul L.H., and Maria José Sousa. 2000. “Biochemical Pathways for the
Production of Flavour Compounds in Cheeses During Ripening: A Review.” Le Lait
V80 (3): pp. 293–324.
Metwalli, Ali A.M, Harmen H.J de Jongh, and Martinus A.J.S van Boekel. 1998. “Heat
Inactivation of Bovine Plasmin.” International Dairy Journal V8 (1) (January 1): pp.
47–56.
Milo, C., and G. A. Reineccius. 1997. “Identification and Quantification of Potent Odorants in
Regular-Fat and Low-Fat Mild Cheddar Cheese.” Journal of Agricultural and Food
Chemistry V45 (9) (September 1): pp. 3590–3594.
Mistry, V.V., and J.-L. Maubois. 2004. “Application of Membrane Separation Technology to
Cheese Production.” In Cheese Chemistry, Physics and Microbiology, Volume 1:pp.
261–VII. Academic Press.
Mistry, V.V., L.E. Metzger, and J.L. Maubois. 1996. “Use of Ultrafiltered Sweet Buttermilk in
the Manufacture of Reduced Fat Cheddar Cheese.” Journal of Dairy Science V79 (7)
(July): pp. 1137–1145.
Moio, Luigi, Dominique Langlois, Patrick Etievant, and Francesco Addeo. 1993. “Powerful
Odorants in Bovine, Ovine, Caprine and Water Buffalo Milk Determined by Means of
86
Gas Chromatography–olfactometry.” Journal of Dairy Research V60 (2): pp. 215–
222.
Molimard, P., and H.E. Spinnler. 1996. “Review: Compounds Involved in the Flavor of
Surface Mold-Ripened Cheeses: Origins and Properties.” Journal of Dairy Science
V79 (2) (February): pp. 169–184.
Mooney, J.S, P.F Fox, A Healy, and J Leaver. 1998. “Identification of the Principal Waterinsoluble Peptides in Cheddar Cheese.” International Dairy Journal V8 (9)
(September): pp. 813–818.
Muir, D.D., J.M. Banks, and E.A. Hunter. 1996. “Sensory Properties of Cheddar Cheese:
Effect of Starter Type and Adjunct.” International Dairy Journal V6 (4) (April): pp.
407–423.
Mulvihill, Donald M., and Patrick F. Fox. 1979. “Proteolytic Specificity of Chymosin on
Bovine Αs1-Casein.” Journal of Dairy Research V46 (4): pp. 641–651.
Naughton, M. A., and F. Sanger. 1961. “Purification and Specificity of Pancreatic Elastase.”
Biochemical Journal V78 (1): pp. 156–163.
Niven, C. F., L. G. Buettner, and James B. Evans. 1954. “Thermal Tolerance Studies on the
Heterofermentative Lactobacilli That Cause Greening of Cured Meat Products.”
Applied Microbiology V2 (1) (January): pp. 26–29.
O’Keeffe, Arthur M., Patrick F. Fox, and Charles Daly. 1978. “Proteolysis in Cheddar Cheese:
Role of Coagulant and Starter Bacteria.” Journal of Dairy Research V45 (3): pp. 465–
477.
O’Keeffe, R. B., P. F. Fox, and C. Daly. 1976. “Contribution of Rennet and Starter Proteases
to Proteolysis in Cheddar Cheese.” Journal of Dairy Research 43 (01): 97–107.
O’Shea, B.A., T. Uniacke-Lowe, and P.F. Fox. 1996. “Objective Assessment of Cheddar
Cheese Quality.” International Dairy Journal V6 (11-12) (November): pp. 1135–
1147.
Olivecrona, T., S. Vilaró, G. Bengtsson-Olivecrona, and P. F Fox. 1992. “Indigenous enzyme
in milk. II. Lipases in milk.” In Advanced Dairy Chemistry 1, 292–310.
Payne, Martin J., and Rohan G. Kroll. 1991. “Methods for the Separation and Concentration
of Bacteria from Foods.” Trends in Food Science & Technology V2 (0): pp. 315–319.
Peterson, S.D., and R.T. Marshall. 1990. “Nonstarter Lactobacilli in Cheddar Cheese: A
Review.” Journal of Dairy Science V73 (6) (June): pp. 1395–1410.
Piraino, Paolo, Eugenio Parente, and Paul L. H. McSweeney. 2004. “Processing of
Chromatographic Data for Chemometric Analysis of Peptide Profiles from Cheese
Extracts:  A Novel Approach.” Journal of Agricultural and Food Chemistry 52 (23)
(November 1): 6904–6911.
Precetti, A.S., M.P. Oria, and S.S. Nielsen. 1997. “Presence in Bovine Milk of Two Protease
Inhibitors of the Plasmin System.” Journal of Dairy Science V80 (8) (August): pp.
1490–1496.
Preininger, Martin, and Werner Grosch. 1994. “Evaluation of Key Odorants of the Neutral
Volatiles of Emmentaler Cheese by the Calculation of Odour Activity Values.” LWT Food Science and Technology V27 (3) (June): pp. 237–244.
Pripp, A.H, Shakeel-Ur-Rehman, P.L.H McSweeney, and P.F Fox. 1999. “Multivariate
Statistical Analysis of Peptide Profiles and Free Amino Acids to Evaluate Effects of
Single-strain Starters on Proteolysis in Miniature Cheddar-type Cheeses.”
International Dairy Journal V9 (7) (July): pp. 473–479.
87
Pripp, Are Hugo, Leszek Stepaniak, and Terje Sørhaug. 2000. “Chemometrical Analysis of
Proteolytic Profiles During Cheese Ripening.” International Dairy Journal V10 (4):
pp. 249–253..
Puchades, R., L. Lemieux, and R.e. Simard. 1989. “Evolution of Free Amino Acids During
the Ripening of Cheddar Cheese Containing Added Lactobacilli Strains.” Journal of
Food Science V54 (4): pp. 885–888.
Qian, M., and G. Reineccius. 2002. “Identification of Aroma Compounds in ParmigianoReggiano Cheese by Gas Chromatography/Olfactometry.” Journal of Dairy Science
V85 (6) (June): pp. 1362–1369.
Rehman, Shakeel-Ur, J.M Banks, E.Y Brechany, D.D Muir, P.L.H McSweeney, and P.F Fox.
2000. “Influence of Ripening Temperature on the Volatiles Profile and Flavour of
Cheddar Cheese Made from Raw or Pasteurised Milk.” International Dairy Journal
V10 (1–2) (January): pp. 55–65.
Richardson, B. C., L. K. Creamer, K. N. Pearce, and R. E. Munford. 1974. “Comparative
Micelle Structure: II. Structure and Composition of Casein Micelles in Ovine and
Caprine Milk as Compared with Those in Bovine Milk.” Journal of Dairy Research
V41 (2): pp. 239–247.
Roy, Denis, Isabelle Mainville, and Francine Mondou. 1997. “Selective Enumeration and
Survival of Bifidobacteria in Fresh Cheese.” International Dairy Journal V7 (12)
(November 21): pp. 785–793.
Sablé, Sophie, and Gilles Cottenceau. 1999. “Current Knowledge of Soft Cheeses Flavor and
Related Compounds.” J. Agric. Food Chem. V47 (12): pp. 4825–4836.
Seefeldt, K.E., and B.C. Weimer. 2000. “Diversity of Sulfur Compound Production in Lactic
Acid Bacteria.” Journal of Dairy Science V83 (12) (December): pp. 2740–2746.
Shakeel-Ur-Rehman, J. M Banks, P. L. H Mcsweeney, and P. F Fox. 2000. “Effect of ripening
temperature on the growth and significance of non-starter lactic acid bacteria in
Cheddar cheese made from raw or pasteurised milk.” International dairy journal V10
(1-2): pp. 45–53.
Singh, T. K, M. A Drake, and K. R Cadwallader. 2003. “Flavor of Cheddar Cheese: A
Chemical and Sensory Perspective.” Comprehensive Reviews in Food Science and
Food Safety V2 (4) (October 1): pp. 166–189.
Singh, T.K., P.F. Fox, P. Højrup, and A. Healy. 1994. “A Scheme for the Fractionation of
Cheese Nitrogen and Identification of Principal Peptides.” International Dairy
Journal V4 (2): pp. 111–122.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1995. “Water-Soluble Peptides in Cheddar
Cheese: Isolation and Identification of Peptides in the Diafiltration Retentate of the
Water-Soluble Fraction.” Journal of Dairy Research V62 (4): pp. 629–640.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy 1997. “Isolation and Identification of Further
Peptides in the Diafiltration Retentate of the Water-soluble Fraction of Cheddar
Cheese.” Journal of Dairy Research V64 (3): pp. 433–443.
Smacchi, Emanuele, and Marco Gobbetti. 1998. “Purification and Characterization of
Cystathionine Γ-lyase from Lactobacillus Fermentum DT41.” FEMS Microbiology
Letters V166 (2): pp. 197–202.
Smit, Gerrit, Bart A. Smit, and Wim J.M. Engels. 2005. “Flavour Formation by Lactic Acid
Bacteria and Biochemical Flavour Profiling of Cheese Products.” FEMS Microbiology
Reviews V29 (3): pp. 591–610.
88
Smith, A. M., and S. Nakai. 1990. “Classification of cheese varieties by multivariate analysis
of HPLC profiles.” Canadian Institute of Food Science and Technology Journal V23
(1): pp. 53–58.
Sousa, M.J, Y Ardö, and P.L.H McSweeney. 2001. “Advances in the Study of Proteolysis
During Cheese Ripening.” International Dairy Journal V11 (4-7) (July 11): pp. 327–
345.
Stuart, Mark R., Lan Szu Chou, and Bart C. Weimer. 1999. “Influence of Carbohydrate
Starvation and Arginine on Culturability and Amino Acid Utilization of Lactococcus
Lactis Subsp. Lactis.” Applied and Environmental Microbiology V65 (2) (February
1): pp. 665–673.
Subramanian, A., W.J. Harper, and L.E. Rodriguez-Saona. 2009. “Cheddar Cheese
Classification Based on Flavor Quality Using a Novel Extraction Method and Fourier
Transform Infrared Spectroscopy.” Journal of Dairy Science V92 (1) (1) (January):
pp. 87–94.
Suriyaphan, Orasa, MaryAnne Drake, X. Q. Chen, and Keith R. Cadwallader. 2001.
“Characteristic Aroma Components of British Farmhouse Cheddar Cheese.” Journal
of Agricultural and Food Chemistry V49 (3) (March 1): pp. 1382–1387.
Tammam, J. D, A. G Williams, J. Noble, D. Lloyd, J. D Tammam, A. G Williams, J. Noble,
and D. Lloyd. 2000. “Amino acid fermentation in non‐starter Lactobacillus spp.
isolated from Cheddar cheese, Amino acid fermentation in non‐starter Lactobacillus
spp. isolated from Cheddar cheese.” Letters in Applied Microbiology, Letters in
Applied Microbiology V30 (5) (May 1): pp. 370–374.
Tanaka, Hidehiko, Nobuyoshi Esaki, and Kenji Soda. 1985. “A Versatile Bacterial Enzyme:
L-methionine Γ-lyase.” Enzyme and Microbial Technology V7 (11) (November): pp.
530–537.
Thomsen, D., and H. Stapelfeldt. 1990. “Yield and quality of Danbo 45+ cheese from milk
heat treated at temperatures between 72 and 95°C.” (44): 39.
Tsakalidou, E., and G. Kalantzopoulos. 1992. “Purification and Partial Characterization of an
Esterase from Lactococcus Lactis Ssp Lactis Strain ACA-DC 127.” Le Lait V72 (6):
pp. 533–543.
Turner, K. W, and T. D Thomas. 1980. “Lactose fermentation in Cheddar cheese and the
effect of salt.” New Zealand Journal of Dairy Science and Technology V15 (3): pp.
265–276.
Upadhyay, V.K., P.L.H. McSweeney, A.A.A. Magboul, and P.F. Fox. 2004. “Proteolysis in
Cheese During Ripening.” In Cheese Chemistry, Physics and Microbiology, Volume
1:pp. 391–433. Academic Press.
Urbach, G. 1995. “Contribution of Lactic Acid Bacteria to Flavour Compound Formation in
Dairy Products.” International Dairy Journal V5 (8): pp. 877–903.
Urbach, Gerda. 1993. “Relations Between Cheese Flavour and Chemical Composition.”
International Dairy Journal V3 (4–6): pp. 389–422.
USDA. 2011. Dairy Products Annual Sumary.
Vazquez-Landaverde, P.A., J.A. Torres, and M.C. Qian. 2006. “Quantification of Trace
Volatile Sulfur Compounds in Milk by Solid-Phase Microextraction and Gas
Chromatography–Pulsed Flame Photometric Detection.” Journal of Dairy Science
V89 (8) (August): pp. 2919–2927.
Verdi, Robert J., and David M. Barbano. 1991. “Properties of Proteases from Milk Somatic
Cells and Blood Leukocytes.” Journal of Dairy Science V74 (7) (July): pp. 2077–
2081.
89
Visser, S. ;Hup. 1983. “Bitter flavour in cheese, 2: model studies on the formation and
degradation of bitter peptides by proteolytic enzymes from calf rennet, starter cells
and starter cell fractions.” Netherlands Milk and Dairy Journal V37 (3): pp. 169–180.
Visser, Servaas. 1993. “Proteolytic Enzymes and Their Relation to Cheese Ripening and
Flavor: An Overview.” Journal of Dairy Science V76 (1) (January): pp. 329–350.
Voigt, Daniela D., François Chevalier, John A. Donaghy, Margaret F. Patterson, Michael C.
Qian, and Alan L. Kelly. 2012. “Effect of High-pressure Treatment of Milk for
Cheese Manufacture on Proteolysis, Lipolysis, Texture and Functionality of Cheddar
Cheese During Ripening.” Innovative Food Science & Emerging Technologies V13
(0) (January): pp. 23–30.
Wallace J.M., and Fox P.F. 1997. “Effect of Adding Free Amino Acids to Cheddar Cheese
Curd on Proteolysis, Flavour and Texture Development.” International Dairy Journal
V7 (2): pp. 157–167.
Walstra, P., T. J Geurts, A. Noomen, A. Jellema, and M. A. J. S. van Boekel. 1999. “Cheese
Varieties.” In Dairy technology: principles of milk properties and processes, pp 651–
708. Marcel Dekker.
Walstra, Pieter, Jan T. M. Wouters, and Tom J. Geurts. 2006. Dairy Science And Technology.
CRC Press.
Weimer, B. C. 2007. Improving the flavour of cheese. CRC Press.
http://www.cabdirect.org/abstracts/20083142333.html.
Weimer, Bart, Kimberly Seefeldt, and Benjamin Dias. 1999. “Sulfur Metabolism in Bacteria
Associated with Cheese.” Antonie Van Leeuwenhoek 76 (1): 247–261.
Wijesundera, C., and L. Drury. “Role of milk fat in production of cheddar cheese flavour
using a fat-substituted cheese model.” Australian journal of dairy technology V54 (1):
pp. 28–35.
Wilkinson, Martin G., Timothy P. Guinee, Daniel M. O’Callaghan, and Patrick F. Fox. 1994.
“Autolysis and Proteolysis in Different Strains of Starter Bacteria During Cheddar
Cheese Ripening.” Journal of Dairy Research V61 (2): pp. 249–262.
De Wit, M., G. Osthoff, B.C. Viljoen, and A. Hugo. 2005. “A Comparative Study of Lipolysis
and Proteolysis in Cheddar Cheese and Yeast-inoculated Cheddar Cheeses During
Ripening.” Enzyme and Microbial Technology V37 (6) (November 1): pp. 606–616.
de Wit, Rudy, and Hans Nieuwenhuijse. 2008. “Kinetic Modelling of the Formation of
Sulphur-containing Flavour Components During Heat-treatment of Milk.”
International Dairy Journal V18 (5) (May): pp. 539–547.
Wong, N. P, R. Ellis, and D. E LaCroix. “Quantitative determination of lactones in Cheddar
cheese.” Journal of Dairy Science V58 (10): pp. 1437–1441.
Yvon, Mireille, Claire Chabanet, and Jean-Pierre Pélissier. 1989. “Solubility of Peptides in
Trichloroacetic Acid (TCA) Solutions Hypothesis on the Precipitation Mechanism.”
International Journal of Peptide and Protein Research V34 (3): pp. 166–176.
Yvon, Mireille, and Liesbeth Rijnen. 2001. “Cheese Flavour Formation by Amino Acid
Catabolism.” International Dairy Journal V11 (4–7) (July 11): pp. 185–201.
Zehentbauer, Gerhard, and G. A. Reineccius. 2002. “Determination of Key Aroma
Components of Cheddar Cheese Using Dynamic Headspace Dilution Assay.” Flavour
and Fragrance Journal V17 (4): pp. 300–305.
90
CHAPTER 3
COMPARISON OF RIPENING CHARACTERISTICS OF CHEDDAR
CHEESE FROM THE SAME COMPANY MANUFACTURED IN DIFFERENT
PRODUCTION PLANTS
Lemus Muñoz, Freddy Mauricio, Qian Michael
Department of Food Science and Technology, Oregon State University
Corvallis Oregon 97331
91
ABSTRACT
The difference in ripening patterns between samples from the same manufacturer but
produced in different plants (location) was studied.
Proteolysis was investigated by a
fractionation scheme, resulting in an insoluble fraction analyzed by urea polyacrylamide gel
electrophoresis (Urea-PAGE), and a soluble fraction which was further investigated through
water soluble nitrogen (WSN), trichloroacetic acid soluble nitrogen (TCA-SN) and
phosphotungstic acid soluble nitrogen (PTA-SN) analyzed by total Kjeldahl nitrogen content
(TKN). Reversed phase high performance liquid chromatography (RP-HPLC) was used to
study the peptide profile of the water soluble fraction. Lipolysis was studied by levels of
individual free fatty acids determined through gas chromatography-flame ionization detection
(GC-FID) after isolation employing solid phase extraction (SPE). Volatile sulfur compounds
were studied using head space solid phase micro-extraction (SPME) coupled with gas
chromatography-pulsed flame photometric detection (PFPD).
The Urea-PAGE method was able to differentiate samples according their age, but it could not
discriminate samples regarding their origin. Nonetheless, measurements of total Kjeldahl
Nitrogen (TKN) of the WSN, TCA-SN, and PTA-SN fractions, and the principal component
analysis of the RP-HPLC peptide profile of the WSN fraction, revealed differences in the rate
and pattern of proteolysis for the samples. Levels of total nitrogen for the WSN, TCA and
PTA fractions increased as cheese aged and were lower for cheeses made in the production
plant CRP (B). From the RP-HPLC analysis data it was developed a PCA model with 3
principal components that accounted for the 80.6% of the variability. This model discriminates
the samples according age and quality, and suggests that the cheese samples from TCCA (A)
plant undergo more or faster proteolysis. FFA profiles reveal significant difference in the
extension of lipolysis, which can be mostly related to variations in manufacturing practices
and indicates that good cheese samples had faster lipolysis. The Volatile Sulfur Compounds
(VSC) analysis showed that cheeses made in the production plant A developed higher
concentrations of H2S, DMS and MeSH, suggesting slower catabolism of sulfur containing
amino acids in cheese made in plant CRP; however H2S did not exhibit a continuous
development as the cheese aged
92
INTRODUCTION
Essentially, cheddar cheese is a casein matrix that contains a balanced mixture of moisture,
peptides, amino acids, free fatty acids, lipids, minerals, microflora and other compounds;
whose complex flavor profile is determined by: 1) variation in the composition and quality of
milk (especially in seasonal dairying countries, where variations of protein and fat levels, and
lactose concentration occurs during the year) and other raw materials; 2) manufacturing
practices; and 3) the extent of biochemical events such as
proteolysis, lipolysis, and
glycolysis occurring during ripening (P. F. Fox et al. 1999). This makes cheese manufacturing
with consistent quality and uniform sensory properties, a really challenging labor. However, it
should be kept in mind that it is expectable that cheeses produced in different regions, or in
different production plants belonging to the same company that follow “identical”
manufacturing procedure, might have unique and distinctive flavor attributes.
The quality of Cheddar cheese is associated to maturity, flavor intensity and texture, and it is
usually assessed by expensive panels of trained people, following a cheese-grader set of
criteria based and dependent on the presence or absence of defects, which besides of being a
time consuming practice, it can result in ambiguous and subjective assessments, consequence
of the different customer and manufacturers preferences from region to region.
Thus, in order to be competitive in an industry producing about 3.3 billion pounds of cheddar
cheese per year (USDA 2011), cheese makers require more reliable standards for classifying
and grading cheese, such as quantitative measurements of compositional parameters involving
instrumental methods and chemical analysis. Hence, an accurate evaluation of flavor quality
will improve the relationship between the final sensory character of the product and the factors
to control it during curd manufacture and ripening.
Because of the capacity to simultaneously monitor many key compounds, the use of
instrumental methods such as gas chromatography (GC), high performance liquid
chromatography (HPLC), and electrophoresis, have allowed the identification of hundreds of
compounds contributing to the characteristic flavor of cheese (Paul L.H. McSweeney and
Sousa 2000), which are mainly separated between sapid and aromatic compounds (T. K Singh,
Drake, and Cadwallader 2003). Nonetheless, an adequate correlation of existing sensory
93
criteria and the chosen measurements is required for predicting, classifying and reproducing
products with equivalent quality. Thus, based on the fact that the volatile fraction contributes
to the aroma while the water soluble fraction play a role in the taste (Aston and Creamer
1986), the discrimination of samples regarding their origin can be done using parameters such
as the profile and abundance of headspace volatiles (Subramanian, Harper, and RodriguezSaona 2009), and degree of proteolysis and lipolysis as indices of ripening (P. F. Fox 1989;
Y.F. Collins, McSweeney, and Wilkinson 2004). The use of proteolysis and lipolysis implies
identification and quantification of fatty acids, amino acids, peptides and soluble nitrogen
among other measurements.
Indeed, the analysis by HPLC have made possible the separation of bitter peptides and the
estimation of the ratio of hydrophobic to hydrophilic peptides, the differentiation between
varieties, between young and mature cheese, and between cheese made from raw and
thermally treated milk (Smith and Nakai 1990; Lau, Barbano, and Rasmussen 1991), which is
complemented by the characterization of
low molecular mass pepetides through
electrophoretic methods (Sousa, Ardö, and McSweeney 2001) and the amino acid composition
analysis by the N-terminals, resulting in the characterization of degradation products (Allan J.
Cliffe, Marks, and Mulholland 1993; A.J. Cliffe, Revell, and Law 1989). In a similar way, gas
chromatography (GC) has allowed the characterization and quantification of more than a
hundred volatile flavor compounds in cheese, which coupled with olfactometry analysis and
headspace extraction techniques such as aroma extract dilution analysis (AEDA ) and solid
phase micro extraction (SPME), have made possible the detection and discrimination of the
potent odorants in different varieties (Preininger and Grosch 1994; Fernández-García 1996;
Qian and Reineccius 2002).
Nonetheless, in order to validate the degree of ripening estimated through chromatography;
experimental controls and/or statistical analysis are required to interpret data, and to establish
a accurate correlation between objective methods, sensory analysis and the classification
according to a particular variable. This makes possible documenting differences in attributes
as consequence of cheese origin, resulting in a better understanding of the process variables
for standardizing the product between different locations of production regions.
94
The objective of this study was to apply objective measurements to characterize the
differences found in cheddar cheese made in different production plants
MATERIALS AND METHODS
CHEESE SAMPLES
Cheeses were manufactured by Tillamok county creamery. Cheeses were made according with
standard protocols in Boardman cheese factory and Tillamok cheese factory. Three blocks of
cheese made from each processing plant were selected randomly from three consecutive
manufacturing days. All cheeses were aged using the same conditions at manufacturer’s
facility, Every 2 months a 2 lb portion was sampled from each block and sent to the lab, where
samples are stored at (-37C) to stop ageing process until analysis is completed
FREE FATTY ACIDS ANALYSIS
Chemicals
Pentanoic acid, heptanoic acid, nonanoic acid, undecanoic acid, and heptadecanoic acid were
used as internal standards, they were purchased from Eastman (Rochester, N.Y., U.S.A).
Butanoic acid, hexanoic acid, octanoic acid, decanoic acid, dodecanoic acid, tetradecanoic
acid, 9-tetradecanoic acid, hexadecanoic acid, 9-hexadecanoic acid, octadecanoic acid, 9octadecanoic acid, 9,12-octadecanoic acid and 6,9,12 octadecanoic acid were used for the
standard stock solution, and were obtained from Aldrich Chemical Co. Inc (Milwaukee,
Wisconsin, U.S.A). Heptane, Isopropanol, Sulfuric acid, anhydrous sodium sulfate,
chloroform, formic acid and diethyl ether were obtained from Fisher.
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Six
grams of this previously freeze-ground cheese, 1 ml of 2N sulphuric acid and 1 ml of internal
standard solution (C5:0, C7:0, C9:0, C11:0 and C17:0 in 1:1 hetpane-isopropanol) were mixed
with 7 grams of anhydrous sodium sulfate and 20 ml of 1:1 diethyl ether- heptane in a 40 ml
amber vial using a sonicator and manual agitation. During sonication, the salt-slurry solution
is initially exposed for 15 minutes, after which each vial is shake vigorously to continue with a
95
second sonication period of 20 minutes. With a glass-Pasteur pipette, the sample extract
(solvent) is transferred to an AccuBOND amino cartridge (Agilent Technologies) conditioned
previously with 10 ml of heptane. After the addition of the sample, the column is washed with
5 ml of 2:1 Chloroform-Isopropanol to remove non volatile triglycerides and phospholipids
using a manifold vacuum chamber. Once the washing step is complete, free fatty acids are
eluted with 5ml of 2% formic acid in diethyl ether, collected in a 20 ml vial and stored in the
freezer until GC analysis.
Chromatography
The analysis was performed using a Hewlett-Packard 5890 gas chromatograph equipped with
a flame ionization detector (FID). Samples were analyzed on a DB-FFAP column (15m x
0.53mm ID, 1 m film thickness; Supelco Wax10, Supelco U.S.A). Injector and detector
temperatures were 250C. Nitrogen was used as carrier gas at a flow rate of 15 ml per minute at
a split ratio of 1 to 1. The oven temperature was programmed for a 2 minutes hold at 60C,
raised to 230C at a rate of 8C per minute with a hold of 20 minutes at 230C.
Quantitative analysis
The levels of free fatty acids concentrations were calculated based on individual peak area
from GC-FID response in comparison to the internal standard peak area, and by using standard
calibration curve of individual free fatty acid using Peak Simple software (SRI instruments,
Torrance, CA). Each experimental value corresponds to the average of the 3 extraction
replicates.
VOLATILE SULFUR COMPOUNDS (VSC’S)
Chemicals
Dimethyl sulfide (DMS) was purchased from TCI America (Portland, OR, U.S.A.); gaseous
methanethiol (MeSH) was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wisconsin,
U.S.A), and a solution was prepared by bubbling the gas into cold methanol; a H2S solution
was prepared by dissolving Na2S.9 H2O (Sigma Co) in acidic water stabilized with citric acid
(pH 3).
96
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Then one
gram of this freshly prepared powder is added to a 20ml vial (formerly flushed with argon),
followed by the addition of 4 ml of 1M citric acid and 20 l of the internal standard solution.
After addition of sample vials were immediately sealed with screw caps with teflon-lined
silicone septa. The vials used in this study were previously deactivated with DMTCS 5%
solution in toluene, toluene, methanol and distillate water.
The volatile sulfur compounds were extracted with an 85 m Carbox-PDMS fiber (Supelco,
Bellefonte, PA, U.S.A.). Prior to use, the fiber was conditioned at 300 C for 90 minutes. The
fiber was then placed into a SPME adapter of a CombiPAL autosampler (CTC analytics AG,
Zwingen, Switzerland) Fitted with a vial heater/agitator. Samples were pre-equilibrated at 500
RPM at 40C for 5 minutes, and the extraction of VSC’s was done at 250 RPM at 40C for 25
minutes. The desorption time was 5 minutes and 30 seconds.
Chromatography
The analysis was performed using a Varian CP-3800 gas chromatograph (Varian, Walnut
Creek, CA, U.S.A.) equipped with a pulsed flame photometric detector (PFPD). The
separation of analytes was made using a DB-FFAP fused silica capillary column (30m, 0.32
mm ID and 1 m film thickness; Agilent, Palo Alto, CA, USA) and nitrogen as carrier gas at
constant flow at 2 ml per minute. The injector temperature was 300 C and it was in the
splitless mode. The oven temperature was programmed for a 3 minutes hold at 35C, raised to
150C at a rate of 10C per minute, held for 5 minute, and then heated to 220C at a rate of 20C
per minute with a final hold of 3 minutes. The PFPD was held at 300 C and 450 V with the
following flow rates: Air 1 at 17 ml per min, H2 at 14 ml per min, and Air 2 at 10 ml/min. The
detector response signal was integrated using the software Star Workstation 6.2, Varian)
Quantitative analysis
Matrix effect
97
In order to retain the matrix effect during the construction of the calibration curves, cheese
powder from the “youngest sample” is used. It is de-volatilized by exposure to room
conditions in a hood for 2 hours. Then 1 gram of powder is added to 4 ml of 1M citric acid in
a 20 ml vial and exposed to a 50C water bath for 30 mins, prior to the addition of standards
and internal standard solutions.
Sulfur standards and internal standard preparation
Two internal standards were used for the quantification of VSC’s: ethyl methyl sulfide (EMS)
for H2S, MeSH and DMS, and isopropyl disulfide (IsoProDS) for DMDS and DMTS. The
concentration of the internal standard solution was 500 ppm EMS and 500 ppm IsoProDS in
methanol. Calibration curves were constructed by spiking cheese samples with a range of
known concentrations of H2S, MeSH and DMS. Hydrogen sulfide (H2S) was prepared by
dissolving Na2S.9 H2O in acidic water (pH = 3). Different concentrations of sodium sulfide
solutions were made, and the concentrations of H2S were calculated based on the amounts of
salt added to the matrix. A standard solution of 100 ppm of DMS was individually prepared in
cooled methanol (-15C), and dilutions were made with cooled methanol at the same
temperature. The mesh standard was prepared as following: 1) newly deactivated, recently
flushed with argon, and cooled vials were used; 2) The original standard solution was made by
bubbling pure MeSH into cooled methanol; 3) Dilutions were prepared by taking aliquots
from the original solution contained in a sealed vial, through the teflon-lined silicone septa by
using a syringe. And then injecting the aliquots into new sealed vials containing proportional
amount of cooled methanol through the septa; 4) 1 gr of devolatilized cheese is added to a
recently flushed vial (argon was used), which is immediately flushed again; 5) simultaneous
argon flushing and addition of 4ml of “free” dissolved oxygen-1M citric acid solution and
quick sealing of the vials; 6) Addition of 20 l of internal standard and MeSH standard
through septa. The identification of target compounds was made by comparing retention times
with those of pure standards. Ratios of the square root of the standard area to the
corresponding square root of the internal standard area were plotted Vs concentration ratios to
determine the relation between the response and concentration for the unknowns. Triplicate
analysis was performed for all samples
98
PROTEOLYSIS
Chemicals
Sulfuric acid was purchased from Fisher Scientific International Inc. (Pittsburgh, PA, U.S.A.);
Trichloroacetic acid was purchased from Alfa Aesar (Royston, UK); And phosphotungstic
acid was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wis, U.S.A)
Sample preparation and fractionation
From each 2lb block of cheese, 60 grams are blended with 120 ml of distilled water pre-heated
to 55C. The mixture is blended for 5 minutes and the homogenate is incubated at 55C for 1
hour. Then the pH is adjusted to 4.6 with 1M HCl and the mixture is centrifuged at 3000g for
30 minutes at 4C. Suspension and supernatant were filtered thoroughly 3 times through glass
wool. The filtrate was safe at -20C for macro blog digestion method analysis, and RP-HPLC
analysis. The insoluble pellet was frozen at -20C for further Urea-PAGE gel electrophoresis
analysis.
The trichloroacetic acid soluble nitrogen fraction (TCA-SN) was prepared by the addition of
25 ml of pH 4.6 soluble fraction (WSN) to 25 ml of 24% trichloroacetic acid solution. Then
the mixture is equilibrated for 2 hours at room temperature and filtered through filter paper
Whatman No 40 before macro blog digestion method analysis.
For the phosphotungstic acid soluble nitrogen fraction (PTA-SN), 10 ml of WSN are added to
7 ml of 3.95 M H2SO4 and 3 ml of 33% phosphotungstic acid solution. Then the mixture is
equilibrated overnight at 4C and filtered through filter paper Whatman No 40 before macro
blog digestion method analysis.
Duplicate analysis was performed for all samples.
Macro blog digestion (Kjeldahl Digestion)
From the fractions collected an aliquot (2 ml for the Water soluble fraction, 1ml for TCA-SN
and 1 ml for PTA-SN) is added into a 70 ml Kjeldahl Digestion flask with 10 ml of H2SO4 and
the catalyst pellet containing 0,075 and 1,5 grams of mercuric oxide and potassium sulfate
respectively. The mixture is warmed to 150 C and hold for 1 hour, then heated to 250 C and
99
hold for 1 hour, and finally heated to 350 C and hold for 2 hours. After digestion the sample is
cooled down overnight to room temperature, and diluted with distillate water to 70 ml,
followed by a gentile agitation. Then a 5 ml aliquot is used to determine the nitrogen content
by a rapid flow analyzer FOSS II.
Reversed phase High performance liquid chromatography analysis
The RP-HPLC analysis was performed using a Shimadzu 6 series liquid chromatograph
(Shimadzu scientific instruments, Kyoto Japan), consisting of an autosampler, 2 pumps, a
multi-wavelength spectrophotometer and a controller unit. It was used a nucleosil RP-8
analytical column (250x 4mm, 5 m particle size, 300 A pore size) and a guard column (4.6
x10 mm) from waters (Milford, MA, U.S.A.). The mobile phase consists of solvent A (0.1%
TFA in deionized and vacuum filtered water) and solvent B (0.1% TFA in acetonitrile). The
elution was monitored at 214nm. The following gradient elution was performed: 1) 100%
solvent A for 5 minutes followed by a linear gradient to 55% solvent B (v/v); 2) elution at
55% solvent B for 6 minutes followed by a linear gradient to 60%; 3) elution at 60% solvent B
for 3 minutes; 4) The column is washed using 95% solvent B during 5 minutes; 5) the column
is equilibrated using 100% solvent A during 10 minutes. The sample (WSN fraction) was
dissolved in solvent A (10 mg per ml) and then micro-centrifuged at 14000 RPM for 10
minutes. An aliquot of 40 l from the extract was injected to a flow rate of 0.75 ml per min.
Electrophoresis
Samples of the water-insoluble nitrogen fraction were dry frozen prior to analysis. Samples
were dissolved in a buffer (0.75 g tris, (hydroxymethyl) methylamine, 49 gr urea and 0.4 ml
concentrated HCl, 0.7 ml 2-mercaptoethanol and 0.15 gr bromophenol blue, dissolved to
100ml) and hold at 50C for 40 min. Urea-polyacrylamide gel electrophoresis (urea-PAGE)
was carried out using a Protean II xi cell vertical slab unit (Bio-Rad Laboratories ltd., Hemel
Hempstead, Herts, UK). Urea-PAGE gels (12.5%) were prepared and run according to the
method or Ardö (1999). Reagents used were obtained by Sigma-Aldrich, Inc and Fisher
Scientific.
100
STATISTICAL ANALYSIS
A two-way analysis of variance (ANOVA) on data was carried out using a general linear
model procedure with Turkey’s pair wise comparison at 95% confidence level, using the
package Minitab 15 (minitab Ltda., Coventry, UK).
RESULTS AND DICUSSION
FREE FATTY ACIDS FFA
It is important to keep in mind that levels of lipolysis measured as FFA released are
considered to be moderate for Cheddar cheese (P.L.H. McSweeney et al. 1993; P. F. Fox et al.
1999; Paul L.H. McSweeney and Sousa 2000), which is the reason why excessive lipolysis is
undesirable and may be considered as rancid by some consumers (Yvonne F. Collins,
McSweeney, and Wilkinson 2003).
Due to their considerably lower perception thresholds (Molimard and Spinnler 1996), the most
important FFA contributing direct and indirectly to the background ofCheddar cheese flavor
are those of short (C4:0 to c8:0) and medium (C10:0 to C12:0) chain.Thus, it is relevant to
point out that evident differences were found between the two types of samples assessed.
Also, noticeable differences were found for long chain fatty acids, and in spite of their low
contribution to the overall flavor of Cheddar cheese, they are still a good index of the degree
of lipolysis. In addition it can be seen in the figures 22 that the slope of the
lines
corresponding to short chain fatty acids in these graphs is steeper, which might indicates that
lipolysis of triglycerides containing short chain fatty acids could be faster. This makes sense
since enzymes have better access to these substrates that are usually located at the sn-1 and sn3 position (Balcão and Malcata 1998). Nonetheless, in this case it is difficult to suggests what
could be the most important enzymatic activity between lipases and esterases, since lipolytic
enzymes are specific for the outer ester bonds of tri- or diacylglycerides (Deeth and Touch
2000; Metwalli, de Jongh, and van Boekel 1998) and it has been reported that butanoic, and
other short and medium chain fatty acids are preferentially released by lipolytic activity (Bills
and Day 1964; Chavarri et al. 1997; Yvonne F. Collins, McSweeney, and Wilkinson 2003).
However, based on the fact that the most important lipolytic activity is provided by LAB
101
enzymes, it is not possible to tell whether lipases or esterases influence more the lipolysis in
these samples without a study of specificity which takes into account the bacteria strain.
As it is mentioned above C4:0 and C6:0 were the FFA with highest increment whereas C18:0,
C14:1, C16:1 and C18:3 were the FFA with the lowest raise during ripening. This is explain
by the accessibility of lipases to this substrates, since all of them are located at the sn-1 and sn3 position (Balcão and Malcata 1998). In addition, it can be seen in figure 21, that the most
dominant FFA were C14:0, C16:0, C18:0 and C18:1; however, in spite of this quantitative
importance, this is related to the fact that these are the most abundant FFA in milk (Yvonne F.
Collins, McSweeney, and Wilkinson 2003). And finally the most important observation is that
lipolysis is faster for cheeses made in the TCCA plant, which is not easy to explain due to it
might be related to many factors, and it is evident that these two facilities are not following the
same procedures or standardization of raw materials. Some of these factors are: 1) differences
in cell viability and autolysis of the starter strain; 2) higher activity of the lipoprotein lipase
LPL as consequence of subtle differences in the preparation of cheese milk such as heating or
homogenization protocols; 3) Differences in the control of relative humidity and temperature
during ripening of curds, which could proportionate better conditions for NSLAB growth; and
4) Differences during salting, which for sure could have inhibitory effect in different zones of
the original blocks, because LAB enzymes are really sensitive to the salt in moisture content
(Gripon et al. 1991; P. F. Fox and Stepaniak 1993).
Figure 21 FFA chromatogram TCCA Vs CRP
102
Figure 22 Development of individual FFA for TCCA and CRP cheese
103
Volatile Sulfur Compounds VSC’s
One type of the distinctive aromas of cheddar cheese results from the decomposition of sulfur
containing amino acids such as cysteine and methionine. Indeed it has been reported that
volatile sulfur compounds (VSC’s) correlate with good Cheddar cheese flavor in spite of their
individual attributes described as garlic, onion, cabbage and skunk (D. J. Manning, Chapman,
and Hosking 1976; B. Weimer, Seefeldt, and Dias 1999). Furthermore, it has been stated that
the most important contributors are H2S, MeSH, and DMS, and contrary compounds such as
104
DMDS, DMTS, 3-methylthiopropionaldehyde, Carbonyl sulfide, carbon disulfide and
dimethyl sulfone do not have a significant odor activity (H. M. Burbank and Qian 2005).
By preventing the development of artifacts through a thorough de-activation work on vials and
injection ports (liner), it is possible to suggest that the results of this work confirm that
hydrogen sulphide (H2S), carbon disulphide (CS2), methanethiol (MeSH), and dimethyl
sulphide (DMS) are the only volatile sulfur compounds formed during cheese ripening, and
that compounds such as dimethyl disulphide (DMDS) and dimethyl trisulphide (DMTS) are
decomposition or oxidation products from MeSH once the samples are exposed to
environments where considerable amounts of oxygen are present.
Accounting with the facts mentioned above, only the development of MeSH, DMS and H2S
will be addressed in this work, thus only calibration curves for these compounds were
calculated. However, good linear correlation coefficients were attained for H2S and DMS.
Unfortunately, in the case of MeSH it was not possible to achieve a satisfying calibration
curve free from DMDS and DMTS, whereby the interpretation of results for this compound
was done based on the area ratio with respect to the internal standard EMS.
The figures 25 and 26, revealed that only MeSH and DMS had a steady development during
ripening, and that its concentration increased as cheese aged in comparison to H2S, which did
not exhibit a regular development pattern in figure 23. Nonetheless, these figures also
demonstrate that there are not significant and noticeable differences of the sulfur attributes
related to the origin of the samples.
Although it was not found in this work, due to the initial increment of the H2S principal
precursor, cysteine, product from the denaturation and incorporation of β-lactoglobulins to
casein micelles (which have a limited amount of this amino acid) as consequence of the heat
treatment of cheese milk, it was expected that H2S showed a rising tendency along the
maturation of samples. However, nothing was visible despite of the possible degradation and
conversion of sulfydryl groups (Fennema and Damodaran 1996) through α- and/or βelimination reaction of cysteine by enzymes such as Cystathionine β- and γ-lyase (found in
brevibacteria and bacilli, potentially NSLAB) resulting in hydrogen sulfide formation (B.
105
Weimer, Seefeldt, and Dias 1999; Seefeldt and Weimer 2000), which is a mechanism still
unclear and not well understood. Moreover, it can be seen that H2S is an important contributor
to the attributes of cheddar cheese based on the concentration found in this work and its odor
threshold, 10 ppb in water (Rychlik et al. 1998), resulting in a relevant odor activity value. In
addition it is possible to notice that in spite of the lack of a trend of generation, the samples
from the TCCA plant had a slightly higher H2S concentrations than those for the CRP plant
samples, which could be mainly explained by 1) subtle differences in the heat treatment of
cheese milk between facilities that promote incorporation of β-lactoglobulins to casein
micelles, 2) differences in the NSLAB microflora or 3) definitively lack of standardization for
the raw milk
Unfortunately, in this work methanethiol could not being confirm as a potent odorant in
cheese due to the lack of effectiveness in constructing a calibration curve free of its oxidation
products DMDS and DMTS. Nonetheless, it can be seen that in both cases the amount of
MeSH increased during ripening time, which is in agreement with Urbach 1995. This might be
attributed to enzymatic reaction provided by secondary microflora rather than by chemical
reaction. And as a matter of fact, regardless it has been reported that MeSH can be obtained in
chemical reduced cheese slurries made without starter cultures from the chemical
decomposition of methathione (Donald J. Manning and Moore 1979; Green and Manning
1982), it is more likely that MeSH can be formed by a enzymatic process, which may require
less activation energy, based on the fact that it is a catalytic reaction by nature (Alting et al.
1995; Smacchi and Gobbetti 1998; Dias and Weimer 1998). The enzymatic formation of
MeSH could be via a single pathway catalysed by the catabolism of L-methionine by Lmethionine γ-lyase (Tanaka, Esaki, and Soda 1985) or cystathionine β or γ-lyase (Alting et al.
1995), or via two-step pathways during transamination of L-methionine in the presence of aketo glutarate to form a-keto-γ-methylthiobutyrate (KMTB) (Gao, Mooberry, and Steele 1998;
Yvon and Rijnen 2001; Amarita et al. 2001), which can then be further broken down
enzymatically to form 3-methylthiopropionaldehyde and MeSH. However, these results show
no significant difference between samples from these two plants regarding this attribute,
which it is not clear or easy to explain. In addition, because of the limited content of sulfur
containing amino acids in caseins, it seems possible that the preparation of the cheese milk
could be very similar or at least not drastically different since higher concentration could be
106
expected from more inclusion of whey protein into the cheese curd as consecuence of the heat
treatment of milk, which is not what happened
Something similar to what is described above occurred in the case of DMS. The results in
figure 26 show that although this compound is usually present in raw and heat treated milk
(Datta et al. 2002; Lee et al. 2007), its concentration increased during the ripening stage, and
slightly higher amounts were found for the TCCA samples. Indeed, the results confirm that
this is an important contributor to the aroma of cheddar cheese (Milo and Reineccius 1997)
since its concentration varied from 5ppm to 30 ppm and its odor threshold in water is 2ppm
(Rychlik 1998), which are amounts comparable to the ones obtained by Burbank and Qian,
2008.
On the other hand, there is no proved and/or clear mechanism for the generation of DMS
during cheese ripening; nonetheless, it has been stated that DMS is product of the metabolism
of propioni-bacteria, present in milk microflora (Baer and Ryba 1992), and formed from
methionine (Curioni and Bosset 2002). Additionally it is known that methionine is mainly
present in β-lactoglobulins and integrated to caseins after thermal denaturation (Fennema and
Damodaran 1996; Datta et al. 2002), which provides the sulfhydryl group required for DMS
generation. Moreover, it can be seen in the figure 26 that there is no significant difference
between samples from these two production plants, however, the concentration of samples
from the TCCA plant are slightly higher, which is really hard to explain and is most likely
attributed to difference in activity of NSLAB. On the one hand, due to the heat treatment of
cheese milk, DMS is most likely formed via protein-bound methionine and the formation of
DMS from methionine required more energy than the formation of MeSH. This means that the
actual reaction is probably way more complicated than that proposed by R. de Wit and
Nieuwenhuijse 2008, which involves the possibility of oxidation of MeSH into DMS and
H2S. On the other hand, the production of DMS by catabolic reaction of methionine during
ripening involves both non-enzymatic and enzymatic decomposition of Sulfonium salts
resulting from the catabolism of methionine or cysteine such as a-keto-γ-methyl thiobutyrate
(KMTB) (Gao, Mooberry, and Steele 1998; Yvon and Rijnen 2001; Amarita et al. 2001), Smethyl thioacetate, S- methyl thiopropionate, S- methyl thiobutyrate, and maybe Smethylmethionine, which can be used as substrate by a wide range of enzymes resulting in
107
DMS formation (Bentley and Chasteen 2004). In addition, there is a possibility that MeSH can
be further transformed to DMS by a methyl transfer reactions involving thiol transferases and
lyases for sulfur-containing amino acids provided by secondary microflora such as
Brevibacterium linens (Dias and Weimer 1998), different strains of Lactococcus lactis subsp.
cremoris (Alting et al. 1995), Lactococcus lactis subsp. lactis, Lacto bacillus sp.,
Propionibacterium shermanii and/or the yeast Geotrichum candidum and Kluyveromyces
lactis (K. Arfi et al. 2002)
Figure 23 VSC's chromatogram for TCCA cheese
Figure 24 Development of H2S for TCCA Vs CRP
cheese
Figure 25Development of MeSH for TCCA Vs CRP
cheese
108
Figure 26 Development of DMS for TCCA Vs CRP
cheese
Effect of treatment on Proteolysis
Based on the results of this work, it is possible to state that measurements of Total Kjeldahl
Nitrogen (TKN) and RP-HPLC revealed clear differences in the rate and pattern of
proteolysis. Unfortunately, the results of the Urea-PAGE test are not conclusive to
discriminate the origin or quality of the samples.
Soluble Nitrogen Fractions and TKN
Following a fractionation scheme in which the extractability of nitrogen compounds depend
on pH (Ardö et al. 1999; Sousa, Ardö, and McSweeney 2001;Voigt et al. 2012), it was
possible to appreciate how nitrogen concentrations increased during time, and was different
regarding to the origin of the samples.
In figures 27 and 28, it can be seen that the nitrogen levels for the WSN fraction were the
highest in comparison to the other fraction, displaying values 4 to 5 times higher than the
TCA-SN fraction and 8 to 18 times higher in comparison to those of the PTA-SN fraction. In
addition, by the end of the observation period the nitrogen content values were at least twice
those at the beginning.
With regard to the levels of WSN, it can be seen that during the first 6 months of maturation,
the nitrogen content is similar for both types of samples (TCCA and CRP); however, after this
period the difference between samples became visible, showing higher values for those
109
corresponding to the TCCA plant. Thus, as it is suggested in the work of Bansal, Piraino, and
McSweeney 2009, using this fraction as a reliable index of primary proteolysis, which is
related to the remaining rennet activity and indigenous milk proteinases (S. Visser 1993; Paul
L. H. McSweeney et al. 1994), it may be possible to focus the attention on variables in the
process that are capable to alter the performance of this enzymes such as, acid development,
temperature profiles, salt uptake and diffuse, and syneresis,
The levels of TCA-SN in figure 30 also illustrate differences in the proteolysis development
between samples from these two plants. In a similar way to the WSN fraction, this one results
in final values that are approximately 3 times higher than those at the beginning of the
observation, showing higher values for the samples of the TCCA plant during the ripening
stage, which are slightly lower values but still comparable to those found by (Voigt et al.
2012).
It is well known that in this fraction the higher the concentration of TCA, the lower the
number of soluble peptides. Therefore, it is probable that a 12%TCA solution can be rich in
medium sized and small peptides, and amino acids with low and medium hydrophobicity
(Kuchroo & Fox 1982), from the N-terminal half of αs1-casein (Tanoj K. Singh, Fox, and
Healy 1997), the N-terminal half of β-casein; most of them products of the action of starter
enzymes (R. B. O’Keeffe, Fox, and Daly 1976; A. M. O’Keeffe, Fox, and Daly 1978), and
endopeptidases from products of Chymosin or Plasmin. Whereby, it would be possible to
attribute this difference to the activity of the starter enzymes. Indeed, the ability to degrade
peptides from the action of Chymosin and Plasmin in cheese by LAB and NSLAB is tightly
determined by the right combination of growth conditions, enzymes activity, ability to lyse
and ripening conditions. Thus water activity and pH determine the survival and growth of
microorganisms in cheese and indirectly their enzyme activity. However, the difference in
substrate availability related to the activity of Plasmin and rennet is still a possibility.As a
matter of fact, much of the work in lab is focused on maximizing cell mass starter activity and
stability during storage rather than in proteolytic activity. Another factor that can be taking
into account could be the lack of reproducible starter performance, related to undefined mixed
strain cultures selected from original natural cultures rather than defined mixtures of pure
characterized strains.
110
The PTA-SN fraction is mainly composed by very small peptides (<15 kDa) and amino acids
of approximately 600 Da (Aston and Dulley 1982), and had a similar trend to that observed in
the results for last fractions, where the values indicate a faster secondary proteolysis for the
TCCA samples. Moreover, it is possible to appreciate that in comparison to the other fractions,
the increment of nitrogen content was more steadfast, displaying final values for the TCCA
samples that were 2.5 higher than those at the beginning. Alike the TCA fraction, these results
could be related to non-reproducible processing variables between the two plants. In addition,
taking in to account that the PTA-SN fraction is a index of secondary proteolysis, the results
for this fraction can be pointed to dissimilar conditions of humidity and temperature in the
maturation rooms for these two plants, or simply a different starter culture.and cheese
microflora
From the results of these fractions, it is reasonable to affirm that the water based fractionation
scheme used is sufficient to extract the majority of water soluble peptides to establish a fair
comparison between products of these two plants. And the most probable explanation to the
difference in the proteolysis is found in dissimilar ripening conditions, and non-reproducible
manufacturing practices.
Figure 27 TKN fractions TCCA cheese
111
Figure 28 TKN fractions CRP cheese
Figure 29 WSN TCCA Vs CRP cheese
Figure 30 TCA-SN TCCA Vs CRP cheese
112
Figure 31 PTA-SN TCCA Vs CRP cheese
Peptide analysis by RP-HPLC
Another index of proteolysis to evaluate the difference between the ripening of samples from
different origin is their peptide profile estimated by RP-HPLC. In this work, it was used a
principal component analysis (PCA) to interpret the raw data from chromatograms based on
Piraino, Parente, and McSweeney, 2004; and Benfeldt and Sørensen, 2001 works; where the
complexity of the profiles containing more than 70 peaks, is initially reduced by establishing a
set of elution time intervals (for which the area of each one is expressed as percentage of the
total area of the chromatogram), and then the variability is measured and analyzed through the
variation of the contributing eigenvalues in the correlation matrix for the principal components
of the resulting model.
A comparison of the ripening characteristics of these Cheddar cheese samples was set up by
randomized block design with three replicates. Visible differences were supported by the PCA
analysis, which resulted in a model with three principal components (PCs) that explain the
80.6% of the variability of the data, and after comparing in pairs the score plots for any
combination of the principal components, only the score plot for the PC1 Vs PC2 revealed a
correlation. These two principal components explain the 70.8% of the variability of the data.
In figures 33 and 34 it can be seen that PC1 differentiate the samples according to their age,
while PC2 represent mainly the contribution from the origin to the variation. Therefore, the
points in these figures displaying higher scores indicate samples with longer ripening that
113
correspond to the TCCA plant, which suggest that proteolysis is slower for cheeses made in
the CRP plant.
The projection of the eluting intervals on the PC1 and PC2 is shown in figure 35, and it
presents the correlation between variables and their effect on the amount of peptides eluting
within certain retention times. The segments 12-20, 20-25, and 35-40 have high scores for the
PC1, which suggest that the amount of peptides eluting in this zones increase during ripening.
In contrast the segments 25-30, 30-35 40-45, 45-50, 50-55, 55-60 and 60-65 have lower
scores, indicating that the amount of peptides in these segments reduces over time. On the
other hand, the segments, 30-35, 55-60 and 60-65 have values close to 0 in the PC2, which
suggest that the peptides eluting in this intervals have no variability as consequence of their
origin. In a different way, the segments 12-20, 20-25, 25-30, 45-50 and 50-55 show low scores
for the PC2, which indicates that the amount of peptides in these zones is smaller for the
samples that correspond to the CRP plant, while the segments 35-40 and 40-45 have high
scores for PC2, displaying a higher concentration of peptides in these intervals for the chesses
made in the TCCA plant.
Regarding the intervals 12-20 and 20-25, based on the works of Tove M. I. E. Christensen,
Kristiansen, and Madsen 1989; T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1995;
Tanoj K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox 1998, it is possible
to say that these sections of the chromatogram are mainly composed by peptides, which are
products from the hydrolysis of the αs1-casein. Indeed, besides to para-κ-caseins, these parts
of the chromatogram are dominated by the peptides αs1-CN(f1-9) and αs1-CN(f1-13), which
are accumulated during cheese ripening and are originated from the hydrolysis of the peptide
αs1-CN(f1-23) by the starter cell-envelope proteinase (CEP). This last peptide is result of the
cleavage of the bond Phe23-Phe24 of αs1-caseins by Chymosin during the early stages of
proteolysis. In addition, other peptides that constitute these segments and are product of the
action of action of CEP and amino peptidase on the peptide αs1-CN (f1-23) are αs1-CN (f18), αs1-CN(f8-23), αs1-CN(f9-23), αs1-CN(f14-23), and the N terminal residues αs1-CN(f10?), αs1-CN(f17-?), αs1-CN(f18-?) and αs1-CN(f11-?). Other peptides in these segments with
different origin are αs1-CN(f25-31), product of the hydrolysis of αs1-CN(f24-199) by
aminopeptidase and/or Chymosin; αs1-CN(f92/93-?), which may involve activity of starter
114
endopeptidases (Pep O, Pep F); And finally αs2-CN(f1-?) and αs2-CN(191-197) that might
involve lactococcal CEP and aminopeptidase activity due to the C terminus of the last peptide.
Therefore, from the kind of peptides in this retention times and the proteolytic systems
involved, it is feasible to attribute the difference between the ripening of the samples from
both plants to the rennet, starter LAB, or to processing variables affecting these systems such
as NaCl (salting), temperature and relative humidity of ripening rooms, and pH at draining,
which in the case of the CRP plant seems like to contribute to the depletion of the enzymatic
activity.
The segments 25-30 and 30-35 got negative loadings for the PC1, which indicates that the
relative amount of peptides eluting in this intervals decrease over time. Additionally it can be
seen that these segment obtained negative loading scores for PC2, which are related to a
slower proteolysis in the samples from the CRP plant. The fact that the amounts of peptides
eluting in the first segment decrease over time has to do with the type of peptides dominating
this zone, which are
αs1-CN, αs1-CN (f24-199) and αs2-CN that tend to decrease as
proteolysis takes place by the activity of Chymosin and CEP, specially over the first two. In
the case of the segment 30-35 min, in spite of the possible accumulation of the peptides αs1CN (f85-91), αs1-CN (f11-?), αs2-CN (f170-?) and αs1-CN (f175-182), the presence of β-CN
seems to strongly influence the amount of relative peptides eluting in this zone, which of
course is expected to decrease over time (in spite of the fact that it is a water insoluble
component and only the 50% of the β-CN are hydrolyzed during ripening in Cheddar cheese)
(Tove M. I. E. Christensen, Kristiansen, and Madsen 1989; T.K. Singh et al. 1994; Tanoj K.
Singh, Fox, and Healy 1995; Tanoj K. Singh, Fox, and Healy 1997; Manuela Fernández,
Singh, and Fox 1998; Sousa, Ardö, and McSweeney 2001). Regarding the scores in PC2, it
might be explained by the fact that the water-soluble peptide profiles are directly related to the
variety, and reflect the specificity of the LAB and NSLAB enzymes (P. F. Fox and
McSweeney 1996). In addition it is obvious the influence of Chymosin on αs1-CN, αs1-CN
(f24-199), whereby it is possible once again to focus the attention on
possible distinct
renneting or starter systems activity, consequence of smaller retention of coagulant activity
due to denaturation by cooking temperature of curds, or because of low moisture level in the
final cheese (P. F. Fox et al. 1999), or because a non optimum pH at draining of curds, which
in addition can cause dissociation of plasmin and plasminogen from micelles. Also the
115
conditions for the LAB and NSLAB growing and survival might be different, probably due to
calibration of instruments.
With respect to the segments 35-40 and 40-45, it will be possible to say that the peptides
eluting in this zone have a hydrophobic character due to the type of mobile phase mixture they
are eluting in. In addition, as it can be seen in the loading plot the relative amount of peptides
in the section 35-40 slightly increased, whereas relative amounts in the section 40-45 decrease
during the ripening. However, both sections had positive scores respect to PC2, specially for
the segment 35-40 min, which suggest that proteolysis related to this type of peptides is faster
for samples from the TCCA plant. From the works of Tove M. I. E. Christensen, Kristiansen,
and Madsen 1989; T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1995; Tanoj K.
Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox 1998, it might be possible to
locate the peptides αs1-CN(f93–?), αs1-CN(f24–30), αs1-CN(f26–32), αs1-CN(f26–34)
resulting from the hydrolysis of the peptide αs1-CN(f24–199) by Chymosin, CEP and
aminopeptidase in the segment 35-40. Also peptides αs2-CN(f204–207), which is a C-terminal
residue and product of lactooccal CEP (Paul L. H. McSweeney et al. 1994), β-CN(f45-52), as
well reported product of hydrolysis by CEP and aminopeptidase, and traces of γ-caseins,
which accumulate during ripening and are mainly present in the retentate of the filtration of
the WSE, correspond to this segment. This again is evidence of the differences mentioned
above.
Regarding the components eluting in the segments from 40 to 65 mins, it can be seen that they
have negative scores respect to PC1, which indicates that their amount decrease during
ripening, and in a similar way these segments, excepting 40-45, have negative scores for the
PC2, suggesting that elements in segment 40-45 undergo a faster proteolysis in the TCCA
plant whereas segments from 45 to 65 have slower proteolysis in the CRP plant. This can be
explain based on the fact that these segments, and fractions studied in earlier works
correspond to the last part of the gradient elution and are mainly composed by amino acids.
Therefore it can be expected that once they are in its free form they become precursors of
flavor compounds, and catabolic reactions, resulting in decreasing amounts of components for
these zones. However, it is difficult to explain the reason why proteolysis goes faster in the
segment 40-45 for the samples from TCCA and slower for the rest of the chromatogram for
116
the samples from CRP since there are no real measurements of what type of amino acids are
present in these regions and the subsequent catabolic pathway for their degradation into flavor
compounds such as α-keto acids, aldehydes, carboxylic acids, alcohols, esters and thioesters
Figure 32 Scree plot of TCCA Vs CRP cheese
Figure 33 Score plot of TCCA Vs CRP cheese (age)
117
Figure 34 Score plot of TCCA Vs CRP cheese (Origin)
Figure 35 Loading plot of TCCA Vs CRP cheese
118
Figure 36 Peptide profile CRP cheese (A and B) Vs TCCA (C and D) cheese, for 2 and 14 months
A
B
C
D
Electrophoresis
From the results of the results of the urea-PAGE of the insoluble nitrogen fraction in figures
38, 39 and 40, it can be seen the progressive degradation of αS1-CN into the peptides αS1-CN
(f24-199), αS1-CN (f121-199), αS1-CN (f99-199), and the β-CN into γ-CNs during ripening.
Also it can be seen that apparently the development of peptides from αS1-CN is faster than that
for peptides from β-CN, which can be related to the fact that Chymosin is the starter of
hydrolysis of caseins and only the 50% of β-CN is hydrolyzed. However the primary
proteolysis pattern for both types of samples is similar during progress of ripening, which is
the reason why it is possible to attribute the difference in quality of these cheeses according to
its origin to the secondary proteolysis rather than the primary proteolysis
119
Figure 37 Urea PAGE for ripening of TCCA cheese
β -CN(f106-209)
β -CN(f29-209)
β -CN(f108-209)
β -CN
αs1 -CN
αz1-CN(f24-199)
αs1-CN(f121-199)
2M
4M
6M
10 M
12 M
15 M
120
Figure 38 Urea PAGE for TCCA Vs CRP cheeses (12 m)
TCCA
CRP
CONCLUSION
The present study demonstrates that the use of FFA profile, VSC’s profile, measurement of
the level of the Total Kjeldahl nitrogen for the WSN, TCA-SN and PTA-SN fractions, and the
analysis of the RP-HPLC peptide profile of the WSN fraction by using a PCA, are effective
tools and ripening indices to differentiate Cheddar cheese samples regarding their age and
origin. The urea-PAGE was effective to differentiate samples by their age; nonetheless it is
clear that it is not sensitive enough to detect differences related to the origin of the sample.
In addition, the results of levels of nitrogen for all the 3 fractions analyzed demonstrated that
proteolysis is faster for cheeses made in the TCCA plant. This was supported by the PCA
model obtained which suggest evident differenced in manufacturing practices between the
evaluated facilities. In a similar way, it was proved that lipolysis is slower for cheese produce
121
in the CRP plant, whose samples showed lower levels of individual FFA. The amounts of
DMS, H2S and MeSH showed differences between samples. However, despite difference is
not significant, it is possible to see a trend indicating that the catabolism of sulfur containing
amino acids such as methionine and cysteine can be faster for the cheeses made in TCCA.
Another interesting outcome from this analysis was to point out the possibility that DMDS
and DMTS are artifacts from extraction and separation procedures rather than metabolites
from the ripening of Cheddar cheese.
REFERENCES
Alting, A. C., W. Engels, S. van Schalkwijk, and F. A. Exterkate. 1995. “Purification and
Characterization of Cystathionine (beta)-Lyase from Lactococcus Lactis Subsp.
Cremoris B78 and Its Possible Role in Flavor Development in Cheese.” Applied and
Environmental Microbiology V61 (11) (November 1): pp. 4037–4042.
Amarita, F., T. Requena, G. Taborda, L. Amigo, and C. Pelaez. 2001. “Lactobacillus Casei
and Lactobacillus Plantarum Initiate Catabolism of Methionine by Transamination.”
Journal of Applied Microbiology V90 (6): pp. 971–978.
Aston, J. W., and L. K. Creamer. 1986. “Contribution of the components of the water-soluble
fraction to the flavour of Cheddar cheese.” New Zealand Journal of Dairy Science and
Technology V21 (3): pp. 229–248.
Aston, J. W., and John R. Dulley. 1982. “Cheddar cheese flavor review.” Australian Journal
of Dairy Technology V37 (2): pp. 59–64.
Baer, A., and I. Ryba. 1992. “Serological Identification of Propionibacteria in Milk and
Cheese Samples.” International Dairy Journal V2 (5): pp. 299–310.
Balcão, Victor M., and F.Xavier Malcata. 1998. “Lipase Catalyzed Modification of Milkfat.”
Biotechnology Advances V16 (2) (March): pp. 309–341.
Bansal, N, Paolo Piraino, and P. L. H. McSweeney. 2009. “Determination of Proteolysis in
Cheese.” In Handbook of Dairy Foods Analysis, pp. 405–417. CRC Press.
Benfeldt, Connie, and John Sørensen. 2001. “Heat Treatment of Cheese Milk: Effect on
Proteolysis During Cheese Ripening.” International Dairy Journal V11 (4–7): pp.
567–574.
Bentley, Ronald, and Thomas G. Chasteen. 2004. “Environmental VOSCs––formation and
Degradation of Dimethyl Sulfide, Methanethiol and Related Materials.” Chemosphere
V55 (3): pp. 291–317.
Bills, D.D., and E.A. Day. 1964. “Determination of the Major Free Fatty Acids of Cheddar
Cheese.” Journal of Dairy Science V47 (7): pp. 733–738.
Burbank, Helen M., and Michael C. Qian. 2005. “Volatile Sulfur Compounds in Cheddar
Cheese Determined by Headspace Solid-phase Microextraction and Gas
Chromatograph-pulsed Flame Photometric Detection.” Journal of Chromatography A
V1066 (1–2): pp. 149–157.
Burbank, Helen, and Michael C. Qian. 2008. “Development of Volatile Sulfur Compounds in
Heat-shocked and Pasteurized Milk Cheese.” International Dairy Journal V18 (8)
(August): pp. 811–818.
122
Chavarri, Felisa, Mailo Virto, Celia Martin, Ana I. Nájera, Arantza Santisteban, Luis J. R.
Barrón, and Mertxe De Renobales. 1997. “Determination of Free Fatty Acids in
Cheese: Comparison of Two Analytical Methods.” Journal of Dairy Research V64
(03): pp. 445–452.
Christensen, Tove M. I. E., Kristian R. Kristiansen, and Jesper S. Madsen. 1989. “Proteolysis
in Cheese Investigated by High Performance Liquid Chromatography.” Journal of
Dairy Research V56 (05): pp. 823–828.
Cliffe, A.J., D. Revell, and B.A. Law. 1989. “A Method for the Reverse Phase High
Performance Liquid Chromatography of Peptides from Cheddar Cheese.” Food
Chemistry 34 (2): 147–160.
Cliffe, Allan J., Jonathan D. Marks, and Francis Mulholland. 1993. “Isolation and
Characterization of Non-volatile Flavours from Cheese: Peptide Profile of Flavour
Fractions from Cheddar Cheese, Determined by Reverse-phase High-performance
Liquid Chromatography.” International Dairy Journal V3 (4–6): pp 379–387.
Collins, Yvonne F., Paul L.H. McSweeney, and Martin G. Wilkinson. 2003. “Lipolysis and
Free Fatty Acid Catabolism in Cheese: a Review of Current Knowledge.”
International Dairy Journal V13 (11): pp. 841–866.
Collins, Y.F., P.L.H. McSweeney, and M.G. Wilkinson. 2004. “Lipolysis and Catabolism of
Fatty Acids in Cheese.” In Cheese Chemistry, Physics and Microbiology, Volume
1:373–389. Academic Press.
Curioni, P.M.G., and J.O. Bosset. 2002. “Key Odorants in Various Cheese Types as
Determined by Gas Chromatography-olfactometry.” International Dairy Journal v12
(12): pp. 959–984.
Datta, N., A. J. Elliott, M. L. Perkins, and H. C. Deeth. 2002. “Ultra-high-temperature (UHT)
Treatment of Milk: Comparison of Direct and Indirect Modes of Heating” V57 (3):
pp. 211–227.
Deeth, H. C., and V. Touch. 2000. “Methods for detecting lipase activity in milk and milk
products.” Australian Journal of Dairy Technology V55 (3): pp. 153–168.
Dias, Benjamin, and Bart Weimer. 1998. “Conversion of Methionine to Thiols by Lactococci,
Lactobacilli, and Brevibacteria.” Applied and Environmental Microbiology V64 (9):
pp. 3320–3326.
Farkye, Nana Y., and Patrick F. Fox. 1990. “Observations on Plasmin Activity in Cheese.”
Journal of Dairy Research V57 (03): pp. 413–418.
Fennema, Owen R., and Damodaran. 1996. “Amino Acids, Peptides, and Proteins.” In Food
Chemistry, pp. 322–425. CRC Press.
Fernández, Manuela, Tanoj K. Singh, and Patrick F. Fox. 1998. “Isolation and Identification
of Peptides from the Diafiltration Permeate of the Water-Soluble Fraction of Cheddar
Cheese.” Journal of Agricultural and Food Chemistry V46 (11): pp. 4512–4517.
Fernández-García, Estrella. 1996. “Use of Headspace Sampling in the Quantitative Analysis
of Artisanal Spanish Cheese Aroma.” Journal of Agricultural and Food Chemistry
V44 (7): pp. 1833–1839.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Fox, P.F. 1989. “Proteolysis During Cheese Manufacture and Ripening.” Journal of Dairy
Science V72 (6): pp. 1379–1400.
Fox, P.F., and L. Stepaniak. 1993. “Enzymes in Cheese Technology.” International Dairy
Journal V3 (4–6): pp. 509–530..
123
Fox, P. F., and P. L. H. McSweeney. 1996. “Proteolysis in Cheese During Ripening.” Food
Reviews International V12 (4): pp. 457–509.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Gao, Song, Ed S. Mooberry, and James L. Steele. 1998. “Use of 13C Nuclear Magnetic
Resonance and Gas Chromatography To Examine Methionine Catabolism by
Lactococci.” Applied and Environmental Microbiology V64 (12): pp. 4670–4675.
Green, Margaret L., and Donald J. Manning. 1982. “Development of Texture and Flavour in
Cheese and Other Fermented Products.” Journal of Dairy Research V49 (04): pp.
737–748.
Gripon, Jean-Claude, V Monnet, G Lambert, and M. J. Desmazeaud. 1991. “Microbial
enzymes in cheese ripening.” In Food enzymology, vol 1.
K. Arfi, H.Spinnler, R. Tache, and P. bonnarme. 2002. “Production of Volatile Compounds by
Cheese-ripening Yeasts: Requirement for a Methanethiol Donor for S-methyl
Thioacetate Synthesis by Kluyveromyces Lactis.” Applied Microbiology and
Biotechnology V58 (4): pp. 503–510.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen. 1991. “Influence of
Pasteurization of Milk on Protein Breakdown in Cheddar Cheese During Aging.”
Journal of Dairy Science V74 (3): pp. 727–740.
Lee, Won-Jae, Dattatreya S. Banavara, Joanne E. Hughes, Jason K. Christiansen, James L.
Steele, Jeffery R. Broadbent, and Scott A. Rankin. 2007. “Role of Cystathionine βLyase in Catabolism of Amino Acids to Sulfur Volatiles by Genetic Variants of
Lactobacillus Helveticus CNRZ 32.” Applied and Environmental Microbiology V73
(9): pp. 3034–3039.
Manning, D. J., Helen R. Chapman, and Zena D. Hosking. 1976. “The Production of Sulphur
Compounds in Cheddar Cheese and Their Significance in Flavour Development.”
Journal of Dairy Research V43 (2): pp. 313–320.
Manning, Donald J., and Carolyn Moore. 1979. “Headspace Analysis of Hard Cheeses.”
Journal of Dairy Research V46 (3): pp. 539–545.
McSweeney, P.L.H., P.F. Fox, J.A. Lucey, K.N. Jordan, and T.M. Cogan. 1993. “Contribution
of the Indigenous Microflora to the Maturation of Cheddar Cheese.” International
Dairy Journal V3 (7): pp. 613–634.
McSweeney, Paul L. H., Sylvie Pochet, Patrick F. Fox, and Aine Healy. 1994. “Partial
Identification of Peptides from the Water-Insoluble Fraction of Cheddar Cheese.”
Journal of Dairy Research V61 (4) (04): pp. 587–590.
McSweeney, Paul L.H., and Maria José Sousa. 2000. “Biochemical Pathways for the
Production of Flavour Compounds in Cheeses During Ripening: A Review.” Le Lait
V80 (3): pp. 293–324.
Metwalli, Ali A.M, Harmen H.J de Jongh, and Martinus A.J.S van Boekel. 1998. “Heat
Inactivation of Bovine Plasmin.” International Dairy Journal V8 (1): pp. 47–56.
Milo, C., and G. A. Reineccius. 1997. “Identification and Quantification of Potent Odorants in
Regular-Fat and Low-Fat Mild Cheddar Cheese.” Journal of Agricultural and Food
Chemistry V45 (9) (September 1): pp. 3590–3594.
Molimard, P., and H.E. Spinnler. 1996. “Review: Compounds Involved in the Flavor of
Surface Mold-Ripened Cheeses: Origins and Properties.” Journal of Dairy Science
V79 (2) (February): pp. 169–184.
124
O’Keeffe, Arthur M., Patrick F. Fox, and Charles Daly. 1978. “Proteolysis in Cheddar Cheese:
Role of Coagulant and Starter Bacteria.” Journal of Dairy Research V45 (3): pp. 465–
477..
O’Keeffe, R. B., P. F. Fox, and C. Daly. 1976. “Contribution of Rennet and Starter Proteases
to Proteolysis in Cheddar Cheese.” Journal of Dairy Research 43 (1): 97–107.
Piraino, Paolo, Eugenio Parente, and Paul L. H. McSweeney. 2004. “Processing of
Chromatographic Data for Chemometric Analysis of Peptide Profiles from Cheese
Extracts:  A Novel Approach.” Journal of Agricultural and Food Chemistry 52 (23):
6904–6911.
Preininger, Martin, and Werner Grosch. 1994. “Evaluation of Key Odorants of the Neutral
Volatiles of Emmentaler Cheese by the Calculation of Odour Activity Values.” LWT Food Science and Technology V27 (3): pp. 237–244.
Qian, M., and G. Reineccius. 2002. “Identification of Aroma Compounds in ParmigianoReggiano Cheese by Gas Chromatography/Olfactometry.” Journal of Dairy Science
V85 (6) pp. 1362–1369.
Seefeldt, K.E., and B.C. Weimer. 2000. “Diversity of Sulfur Compound Production in Lactic
Acid Bacteria.” Journal of Dairy Science V83 (12): pp. 2740–2746.
Singh, T. K, M. A Drake, and K. R Cadwallader. 2003. “Flavor of Cheddar Cheese: A
Chemical and Sensory Perspective.” Comprehensive Reviews in Food Science and
Food Safety V2 (4): pp. 166–189.
Singh, T.K., P.F. Fox, P. Højrup, and A. Healy. 1994. “A Scheme for the Fractionation of
Cheese Nitrogen and Identification of Principal Peptides.” International Dairy
Journal V4 (2): pp. 111–122.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1995. “Water-Soluble Peptides in Cheddar
Cheese: Isolation and Identification of Peptides in the Diafiltration Retentate of the
Water-Soluble Fraction.” Journal of Dairy Research V62 (4): pp. 629–640.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1997. “Isolation and Identification of Further
Peptides in the Diafiltration Retentate of the Water-soluble Fraction of Cheddar
Cheese.” Journal of Dairy Research V64 (3): pp. 433–443.
Smacchi, Emanuele, and Marco Gobbetti. 1998. “Purification and Characterization of
Cystathionine Γ-lyase from Lactobacillus Fermentum DT41.” FEMS Microbiology
Letters V166 (2): pp. 197–202.
Smith, A. M., and S. Nakai. 1990. “Classification of cheese varieties by multivariate analysis
of HPLC profiles.” Canadian Institute of Food Science and Technology Journal V23
(1): pp. 53–58.
Sousa, M.J, Y Ardö, and P.L.H McSweeney. 2001. “Advances in the Study of Proteolysis
During Cheese Ripening.” International Dairy Journal V11 (4-7): pp. 327–345.
Subramanian, A., W.J. Harper, and L.E. Rodriguez-Saona. 2009. “Cheddar Cheese
Classification Based on Flavor Quality Using a Novel Extraction Method and Fourier
Transform Infrared Spectroscopy.” Journal of Dairy Science V92 (1): pp. 87–94.
Tanaka, Hidehiko, Nobuyoshi Esaki, and Kenji Soda. 1985. “A Versatile Bacterial Enzyme:
L-methionine Γ-lyase.” Enzyme and Microbial Technology V7 (11) (November): pp.
530–537.
Urbach, G. 1995. “Contribution of Lactic Acid Bacteria to Flavour Compound Formation in
Dairy Products.” International Dairy Journal V5 (8): pp. 877–903.
USDA. 2011. Dairy Products Annual Sumary.
Visser, Servaas. 1993. “Proteolytic Enzymes and Their Relation to Cheese Ripening and
Flavor: An Overview.” Journal of Dairy Science V76 (1): pp. 329–350.
125
Voigt, Daniela D., François Chevalier, John A. Donaghy, Margaret F. Patterson, Michael C.
Qian, and Alan L. Kelly. 2012. “Effect of High-pressure Treatment of Milk for
Cheese Manufacture on Proteolysis, Lipolysis, Texture and Functionality of Cheddar
Cheese During Ripening.” Innovative Food Science & Emerging Technologies V13
(0): pp. 23–30.
Weimer, Bart, Kimberly Seefeldt, and Benjamin Dias. 1999. “Sulfur Metabolism in Bacteria
Associated with Cheese.” Antonie Van Leeuwenhoek 76 (1): 247–261.
de Wit, Rudy, and Hans Nieuwenhuijse. 2008. “Kinetic Modelling of the Formation of
Sulphur-containing Flavour Components During Heat-treatment of Milk.”
International Dairy Journal V18 (5): pp. 539–547.
Yvon, Mireille, and Liesbeth Rijnen. 2001. “Cheese Flavour Formation by Amino Acid
Catabolism.” International Dairy Journal V11 (4–7): pp. 185–201.
126
CHAPTER 4
COMPARISION OF CHEDDAR CHEESE RIPENING MANUFACTURED
WITH AND WITHOUT ADJUNCT CULTURE
Lemus Muñoz, Freddy Mauricio, Qian Michael
Department of Food Science and Technology, Oregon State University
Corvallis Oregon 97331
127
ABSTRACT
The effect of addition of adjunct culture isolated from cheese manufactured at the TCCA plant
to cheese produce in the CRP plant was studied during the ripening stage of samples.
Proteolysis was investigated by a fractionation scheme, resulting in an insoluble fraction
analyzed by urea polyacrylamide gel electrophoresis (Urea-PAGE), and a soluble fraction
which was further studied through water soluble nitrogen (WSN), trichloroacetic acid soluble
nitrogen (TCA-SN) and phosphotungstic acid soluble nitrogen (PTA-SN) analyzed by total
Kjeldahl nitrogen content (TKN). Reversed phase high performance liquid chromatography
(RP-HPLC) was used to study the peptide profile of the water soluble fraction. Lipolysis was
studied by levels of individual free fatty acids determined through gas chromatography-flame
ionization detection (GC-FID) after isolation employing solid phase extraction (SPE). Volatile
sulfur compounds were studied using head space solid phase micro-extraction (SPME)
coupled with gas chromatography-pulsed flame photometric detection (PFPD).
The Urea-PAGE method was able to differentiate samples according their age, but it could not
discriminate samples regarding their treatment. Nonetheless, measurements of total Kjeldahl
Nitrogen (TKN) of the WSN, TCA-SN, and PTA-SN fractions, and the principal component
analysis of the RP-HPLC peptide profile of the WSN fraction, revealed differences in the rate
and pattern of proteolysis for the samples. Levels of total nitrogen for the WSN, TCA and
PTA fractions increased as cheese aged and were lower for cheeses made without adjunct
culture. The principal component analysis of the RP-HPLC data resulted in PCA model with 3
principal components that accounted for the 83.4% of the variability. This model discriminates
the samples according age and treatment, suggesting that samples made with adjunct culture
undergoes more or faster proteolysis. FFA profiles reveal significant difference in the
extension of lipolysis, which was higher or faster for cheese made with adjunct culture. The
Volatile Sulfur Compounds (VSC) analysis showed that cheeses made with adjunct culture
developed higher concentrations of H2S, DMS and MeSH, suggesting slower catabolism of
sulfur containing amino acids in cheese made without adjunct culture.
128
INTRODUCTION
Cheddar cheese is the product of a dehydration process involving 2 stages: 1) preparation of
curd, which is usually done during the first 24 hours, and 2) ripening of curds, which can take
up to 12 months. The development of flavor and texture is the result of three complex
biochemical and microbiological process occurring during ripening such as glycolysis,
lipolysis and proteolysis, whose early products constitute the precursors for the formation of
volatile and non-volatile flavor compounds (Wallace J.M. and Fox P.F. 1997; Paul L.H.
McSweeney and Sousa 2000).
During ripening the major sources of enzymatic activity are the rennet enzymes (pepsin and
Chymosin), indigenous milk enzymes (Plasmin and Cathepsin B and D), and starter lactic acid
bacteria (LAB), non-starter lactic acid bacteria (NSLAB), and adjunct cultures enzymes (P. F.
Fox et al. 1999). This stage is determined mainly by the extent of proteolysis (P. F. Fox 1989),
which is the biochemical event where proteolytic enzymes hydrolyze αs1, αs2, β and κ caseins,
originating large and intermediate peptides that are further hydrolyzed by LAB, NSLAB and
adjunct culture intracellular enzymes into free amino acids (FAA) and other smaller peptides,
which are important contributors to the background of cheese flavor and essential precursors
for deamination, decarboxylation and/or transamination reactions, among others, to produce
volatile and sapid compounds (Allan J. Cliffe, Marks, and Mulholland 1993; P. F. Fox et al.
1999; Paul L.H. McSweeney and Sousa 2000). In a similar way fat plays an important role
during ripening, and besides of being solvent for aromatic and non-volatile compounds, its
degradation through lipolysis results in the release of free fatty acids (FFA), which are flavor
compounds that also can be catabolized into other compounds such as methyl ketones,
alcohols and lactones (Gerda Urbach 1993; P. F. Fox et al. 1999). Regarding glycolysis and
ripening, residual lactose is mainly metabolized to l-lactate, and further transformed to acetate
through oxidation by LAB, NSLA or adjunct culture enzymes; this compound has been found
significant in cheddar cheese flavor (P. F. Fox et al. 1996), and can be a positive attribute or
an off-flavor depending on its concentration.
However, ripening is an expensive and long process that has received significant attention due
to the interest of manufacturers in accelerating this stage in order to reduce cost of production,
and to improve the quality and consistency of attributes of the ready to sell product. This fact
129
has promoted research related to the addition of adjunct cultures to accelerate ripening and
improve flavor (P. F. Fox and Stepaniak 1993; C.N. Lane and Fox 1996b).
Co-starting or adjunct cultures can be nonstarter lactic acid bacteria, consisting mainly of
Lactobacillus sp, or certain yeast species. The dominant NSLAB strains are mesophilic
lactobacilli such as Lactobacillus casei ssp. casei, Lactobacillus paracasei ssp. paracasei,
Lactobacillus plantarum, Lactobacillus curvatus, and Lactobacillus helveticus) (P.L.H.
McSweeney et al. 1993; C.N. Lane and Fox 1996b; M. El Soda, Madkor, and Tong 2000) as
well as other probiotic strains such as Lactobacillusacidophilus 4962, Lb. casei 279,
Bifidobacterium longum 1941, Lb. acidophilus LAFTI® L10 (Shah 2006), that have been
recently studied. While in the case of yeast, the preferred strains to work with are
Debaryomyces hansenii and Yarrowia lipolytica (Ferreira and Viljoen 2003; M. De Wit et al.
2005).
In order to take full advantage of adjunct NSLAB cultures, studies about cell autolysis have
been conducted through the use of attenuated cultures, which has been done by means of
sublethal treatments such as freeze shocking (FS), heat shocking (HS), and spray drying (SD),
to provide more control on the release of intracellular enzymes in to the cheese matrix (M. A.
El Soda 1993; M. El Soda, Madkor, and Tong 2000; Madkor, Tong, and El Soda 2000),
reporting acceleration of the breakdown of peptides and consequently increase in the amount
of amino nitrogen and decrease in bitterness, a considerably enhanced flavor intensity without
affecting cheese composition, and also a reduction in off-flavors (Madkor, Tong, and El Soda
2000; M. El Soda, Madkor, and Tong 2000).
Nevertheless, because of acidification is one of the crucial operations in the manufacture of
cheddar cheese, adjunct cultures should meet certain criteria in order to be considered as
agents for accelerating the ripening process. Therefore, qualities such as: 1) a potent lipolytic
and proteolytic system with high autolytic activity, 2) assimilation of lactose and organic
acids, 3) resistance to high salt concentration, low water activity and low pH, 4) ability to
grow at low temperatures, and 5) in the case of yeast cultures, compatibility with the starter
culture; are desire and should be take into account at the moment of selecting any culture.
130
The objective of this study is to establish a comparison between samples of Cheddar cheese
manufactured in the CRP plant that were made with and without adjunct culture isolated from
cheese produced in the TCCA plant, by monitoring lipolysis and proteolysis using: 1) free
fatty acids profiles, 2) peptides profiles by RP-HPLC, 3) Urea-PAGE, 4) development of
sulfur volatile compounds, and 5) nitrogen content of cheese fraction
MATERIALS AND METHODS
CHEESE SAMPLES
Cheeses were manufactured by Tillamok county creamery. Cheeses were made
according with standard protocols in the CRP plant The adjunct culture included is
unknown and was extracted from good quality cheese produced at the TCCA plant.
One blocks of cheese made from each treatment were selected randomly. All cheeses
were aged using the same conditions at manufacturer’s facility, Every 2 months a 2 lb
portion was sampled from each block and sent to the lab, where samples are stored at
(-37C) to stop ageing process until analysis is completed.
FREE FATTY ACIDS ANALYSIS
Chemicals
Pentanoic acid, heptanoic acid, nonanoic acid, undecanoic acid, and heptadecanoic acid were
used as internal standards, they were purchased from Eastman (Rochester, N.Y., U.S.A).
Butanoic acid, hexanoic acid, octanoic acid, decanoic acid, dodecanoic acid, tetradecanoic
acid, 9-tetradecanoic acid, hexadecanoic acid, 9-hexadecanoic acid, octadecanoic acid, 9octadecanoic acid, 9,12-octadecanoic acid and 6,9,12 octadecanoic acid were used for the
standard stock solution, and were obtained from Aldrich Chemical Co. Inc (Milwaukee,
Wisconsin, U.S.A). Heptane, Isopropanol, Sulfuric acid, anhydrous sodium sulfate,
chloroform, formic acid and diethyl ether were obtained from Fisher.
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Six
131
grams of this previously freeze-ground cheese, 1 ml of 2N sulphuric acid and 1 ml of internal
standard solution (C5:0, C7:0, C9:0, C11:0 and C17:0 in 1:1 hetpane-isopropanol) were mixed
with 7 grams of anhydrous sodium sulfate and 20 ml of 1:1 diethyl ether- heptane in a 40 ml
amber vial using a sonicator and manual agitation. During sonication, the salt-slurry solution
is initially exposed for 15 minutes, after which each vial is shake vigorously to continue with a
second sonication period of 20 minutes. With a glass-Pasteur pipette, the sample extract
(solvent) is transferred to an AccuBOND amino cartridge (Agilent Technologies) conditioned
previously with 10 ml of heptane. After the addition of the sample, the column is washed with
5 ml of 2:1 Chloroform-Isopropanol to remove non volatile triglycerides and phospholipids
using a manifold vacuum chamber. Once the washing step is complete, free fatty acids are
eluted with 5ml of 2% formic acid in diethyl ether, collected in a 20 ml vial and stored in the
freezer until GC analysis.
Chromatography
The analysis was performed using a Hewlett-Packard 5890 gas chromatograph equipped with
a flame ionization detector (FID). Samples were analyzed on a DB-FFAP column (15m x
0.53mm ID, 1 m film thickness; Supelco Wax10, Supelco U.S.A). Injector and detector
temperatures were 250C. Nitrogen was used as carrier gas at a flow rate of 15 ml per minute at
a split ratio of 1 to 1. The oven temperature was programmed for a 2 minutes hold at 60C,
raised to 230C at a rate of 8C per minute with a hold of 20 minutes at 230C.
Quantitative analysis
The levels of free fatty acids concentrations were calculated based on individual peaks areas
from GC-FID response in comparison to the internal standard peaks areas, using standard
calibration curve of individual free fatty acid using Peak Simple software (SRI instruments,
Torrance, CA). Each experimental value corresponds to the average of the 3 extraction
replicates.
VOLATILE SULFUR COMPOUNDS (VSC’S)
Chemicals
Dimethyl sulfide (DMS) was purchased from TCI America (Portland, OR, U.S.A.); gaseous
methanethiol (MeSH) was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wisconsin,
132
U.S.A), and a solution was prepared by bubbling the gas into cold methanol; a H2S solution
was prepared by dissolving Na2S.9 H2O (Sigma Co) in acidic water stabilized with citric acid
(pH 3).
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Then one
gram of this freshly prepared powder is added to a 20ml vial (formerly flushed with argon),
followed by the addition of 4 ml of 1M citric acid and 20 l of the internal standard solution.
After addition of sample vials were immediately sealed with screw caps with teflon-lined
silicone septa. The vials used in this study were previously deactivated to its use with DMTCS
5% solution in toluene, toluene, methanol and distillate water.
The volatile sulfur compounds were extracted with a 85 m Carbox-PDMS fiber (Supelco,
Bellefonte, PA, U.S.A.). Prior to use, the fiber was conditioned at 300 C for 90 minutes. The
fiber was then placed into a SPME adapter of a CombiPAL autosampler (CTC analytics AG,
Zwingen, Switzerland) Fitted with a vial heater/agitator. Samples were pre-equilibrated at 500
RPM at 40C for 5 minutes, and the extraction of VSC’s was done at 250 RPM at 40C for 25
minutes. The desorption time was 5 minutes and 30 seconds.
Chromatography
The analysis was performed using a Varian CP-3800 gas chromatograph (Varian, Walnut
Creek, CA, U.S.A.) equipped with a pulsed flame photometric detector (PFPD). The
separation of analytes was made using a DB-FFAP fused silica capillary column (30m, 0.32
mm ID and 1 m film thickness; Agilent, Palo Alto, CA, USA) and nitrogen as carrier gas at
constant flow at 2 ml per minute. The injector temperature was 300 C and it was in the
splitless mode. The oven temperature was programmed for a 3 minutes hold at 35C, raised to
150C at a rate of 10C per minute, held for 5 minute, and then heated to 220C at a rate of 20C
per minute with a final hold of 3 minutes. The PFPD was held at 300 C and 450 V with the
following flow rates: Air 1 at 17 ml per min, H2 at 14 ml per min, and Air 2 at 10 ml/min. The
detector response signal was integrated using the software Star Workstation 6.2, Varian)
133
Quantitative analysis
Matrix effect
In order to retain the matrix effect during the construction of the calibration curves, cheese
powder from the “youngest sample” is used. It is de-volatilized by exposure to room
conditions in a hood for 2 hours. Then 1 gram of powder is added to 4 ml of 1M citric acid in
a 20 ml vial and exposed to a 50C water bath for 30 mins, prior to the addition of standards
and internal standard solutions.
Sulfur standards and internal standard preparation
Two internal standards were used for the quantification of VSC’s: ethyl methyl sulfide (EMS)
for H2S, MeSH and DMS, and isopropyl disulfide (IsoProDS) for DMDS and DMTS. The
concentration of the internal standard solution was 500 ppm EMS and 500 ppm IsoProDS in
methanol. Calibration curves were constructed by spiking cheese samples with a range of
known concentrations of H2S, MeSH and DMS. Hydrogen sulfide (H2S) was prepared by
dissolving Na2S.9 H2O in acidic water (pH = 3). Different concentrations of sodium sulfide
solutions were made, and the concentrations of H2S were calculated based on the amounts of
salt added to the matrix. A standard solution of 100 ppm of DMS was individually prepared in
cooled methanol (-15C), and dilutions were made with cooled methanol at the same
temperature. The mesh standard was prepared as following: 1) newly deactivated, recently
flushed with argon, and cooled vials were used; 2) The original standard solution was made by
bubbling pure MeSH into cooled methanol; 3) Dilutions were prepared by taking aliquots
from the original solution contained in a sealed vial, through the teflon-lined silicone septa by
using a syringe. And then injecting the aliquots into new sealed vials containing proportional
amount of cooled methanol through the septa; 4) 1 gr of devolatilized cheese is added to a
recently flushed vial (argon was used), which is immediately flushed again; 5) simultaneous
argon flushing and addition of 4ml of “free” dissolved oxygen-1M citric acid solution and
quick sealing of the vials; 6) Addition of 20 l of internal standard and MeSH standard
through septa. The identification of target compounds was made by comparing retention times
with those of pure standards. Ratios of the square root of the standard area to the
corresponding square root of the internal standard area were plotted Vs concentration ratios to
determine the relation between the response and concentration for the unknowns. Triplicate
analysis was performed for all samples
134
PROTEOLYSIS
Chemicals
Sulfuric acid was purchased from Fisher Scientific International Inc. (Pittsburgh, PA, U.S.A.);
Trichloroacetic acid was purchased from Alfa Aesar (Royston, UK); And phosphotungstic
acid was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wis, U.S.A)
Sample preparation and fractionation
From each 2lb block of cheese, 60 grams are blended with 120 ml of distilled water pre-heated
to 55C. The mixture is blended for 5 minutes and the homogenate is incubated at 55C for 1
hour. Then the pH is adjusted to 4.6 with 1M HCl and the mixture is centrifuged at 3000g for
30 minutes at 4C. Suspension and supernatant were filtered thoroughly 3 times through glass
wool. The filtrate was safe at -20C for macro blog digestion method analysis, and RP-HPLC
analysis. The insoluble pellet was frozen at -20C for further Urea-PAGE gel electrophoresis
analysis.
The trichloroacetic acid soluble nitrogen fraction (TCA-SN) was prepared by the addition of
25 ml of pH 4.6 soluble fraction (WSN) to 25 ml of 24% trichloroacetic acid solution. Then
the mixture is equilibrated for 2 hours at room temperature and filtered through filter paper
Whatman No 40 before macro blog digestion method analysis.
For the phosphotungstic acid soluble nitrogen fraction (PTA-SN), 10 ml of WSN are added to
7 ml of 3.95 M H2SO4 and 3 ml of 33% phosphotungstic acid solution. Then the mixture is
equilibrated overnight at 4C and filtered through filter paper Whatman No 40 before macro
blog digestion method analysis.
Duplicate analysis was performed for all samples.
Macro blog digestion (Kjeldahl Digestion)
From the fractions collected an aliquot (2 ml for the Water soluble fraction, 1ml for TCA-SN
and 1 ml for PTA-SN) is added into a 70 ml Kjeldahl Digestion flask with 10 ml of H2SO4 and
the catalyst pellet containing 0,075 and 1,5 grams of mercuric oxide and potassium sulfate
135
respectively. The mixture is warmed to 150 C and hold for 1 hour, then heated to 250 C and
hold for 1 hour, and finally heated to 350 C and hold for 2 hours. After digestion the sample is
cooled down overnight to room temperature, and diluted with distillate water to 70 ml,
followed by a gentile agitation. Then a 5 ml aliquot is used to determine the nitrogen content
by a rapid flow analyzer FOSS II.
Reversed phase High performance liquid chromatography analysis
The RP-HPLC analysis was performed using a Shimadzu 6 series liquid chromatograph
(Shimadzu scientific instruments, Kyoto Japan), consisting of an autosampler, 2 pumps, a
multi-wavelength spectrophotometer and a controller unit. It was used a nucleosil RP-8
analytical column (250x 4mm, 5 m particle size, 300 A pore size) and a guard column (4.6
x10 mm) from waters (Milford, MA, U.S.A.). The mobile phase consists of solvent A (0.1%
TFA in deionized and vacuum filtered water) and solvent B (0.1% TFA in acetonitrile). The
elution was monitored at 214nm. The following gradient elution was performed: 1) 100%
solvent A for 5 minutes followed by a linear gradient to 55% solvent B (v/v); 2) elution at
55% solvent B for 6 minutes followed by a linear gradient to 60%; 3) elution at 60% solvent B
for 3 minutes; 4) The column is washed using 95% solvent B during 5 minutes; 5) the column
is equilibrated using 100% solvent A during 10 minutes. The sample (WSN fraction) was
dissolved in solvent A (10 mg per ml) and then micro-centrifuged at 14000 RPM for 10
minutes. An aliquot of 40 l from the extract was injected to a flow rate of 0.75 ml per min.
Electrophoresis
Samples of the water-insoluble nitrogen fraction were dry frozen prior to analysis. Samples
were dissolved in a buffer (0.75 g tris, (hydroxymethyl) methylamine, 49 gr urea and 0.4 ml
concentrated HCl, 0.7 ml 2-mercaptoethanol and 0.15 gr bromophenol blue, dissolved to
100ml) and hold at 50C for 40 min. Urea-polyacrylamide gel electrophoresis (urea-PAGE)
was carried out using a Protean II xi cell vertical slab unit (Bio-Rad Laboratories ltd., Hemel
Hempstead, Herts, UK). Urea-PAGE gels (12.5%) were prepared and run according to the
method or Ardö (1999). Reagents used were obtained by Sigma-Aldrich, Inc and Fisher
Scientific.
136
STATISTICAL ANALYSIS
A two-way analysis of variance (ANOVA) on data was carried out using a general linear
model procedure with Turkey’s pair wise comparison at 95% confidence level, using the
package Minitab 15 (minitab Ltda., Coventry, UK).
RESULTS AND DICUSSION
FREE FATTY ACIDS FFA
Levels of individual free fatty acids are shown in figures 40. It can be seen in figure 39 that
the most dominant peaks were C14:0, C16:0, C18:0 and C18:1, however, despite these
apparent relevant quantities, they are not important contributors to the Cheddar cheese aroma
because of their high odor threshold. In addition, their high concentration is related to the fact
that these are the most abundant FFA in milk (Banks 1991; Gunstone, Harwood, & Padley
1994), and also it is possible to see in the results that these FFA and C14:1, C16:1, C18:1 and
C18:3 are the ones with the lowest increase over time, which is related to the harder access of
lipases to their active sites. Indeed, most of them are located in the same position than the
short chain fatty acids within the tri and diacylglycerides. On the other hand, the increment for
the short chain fatty acids C4:0 and C6:0 was proportionally the highest and fastest, reaching
at least twice their concentration at the beginning of the observation period, which makes
sense since lipases and esterases have better access to these substrates located at the sn-1 and
sn-3 position of the triacylglicerides.
Also it is possible to appreciate that cheeses made with adjunct culture in general exhibited
higher levels of FFA liberation during ripening in comparison to the control samples. These
results indicate that adjunct cultures (possibly other strains of lactococci or lactobacilli)
contribute to lipolysis, which is in agreement with Madkor, Tong, and El Soda 2000, and in
addition, they have high activity of intracellular lipase upon autolysis. However it is important
to keep in mind that Lactococcus and Lactobacillus spp are considered to be weakly lipolytic
in comparison to species such as Pseudomonas, Acinetobacter and Flavobacterium, and it
should be expected that the degree and pace of lipolysis is related to the autolysis capacity of
the strains selected, which in this case are unknown. Therefore it is important to take into
account that the adjunct culture used could have high rates of autolysis with high level of
137
enzymatic activity, or high autolytic activity and low level of enzymatic activity, or low rates
of autolysis with high level of enzymatic activity, or low autolysis and enzymatic activity (ElSoda, Madkor, and Tong 2000).
Figure 39 Urea PAGE for TCCA Vs CRP cheeses (2 months)
Figure 40 Development of individual FFA for cheese with and without adjunct culture
138
139
Volatile Sulfur Compounds VSC’s
Volatile sulfur compounds correlate with good Cheddar cheese flavor, and the most important
contributors are H2S, MeSH, and DMS. These are product from the decomposition of amino
acids such as cysteine and methionine (B. Weimer, Seefeldt, and Dias 1999; D. J. Manning,
Chapman, and Hosking 1976; H. M. Burbank and Qian 2005).
As it was mentioned earlier in the last two chapters, Hydrogen sulphide (H2S), carbon
disulphide (CS2), methanethiol (MeSH), and dimethyl sulphide (DMS) are the only volatile
sulfur compounds originally found in this work as consequence of biochemical reactions
during ripening. Therefore only their development was discussed; however, because CS2 is not
considered as a relevant contributor to cheese aroma it is not going to be part of the analysis.
Calibration curves were calculated for H2S, DMS and MeSH; nonetheless only for the first
two compounds acceptable linear correlation was achieved, reason why the interpretation of
results for MeSH was done based on the area ratio respect the internal standard EMS.
The development of DMS and MeSH was similar to the last two studies; it was steady during
the observation period, resulting in increasing concentration of these compounds over time. In
the case of H2S there was not an evident development pattern, but a clear higher concentration
for those samples made with adjunct culture was observed.
Most of adjunct cultures for Cheddar cheese are NSLAB consisting mainly of lactobacillus
sp.; however, other strains of lactoccus lactis sp. and Brevibacterium linens are included in
this group (C. M. Lynch et al. 1996; C. M. Lynch et al. 1999; P. Fox, McSweeney, and Lynch
140
1998; B. Weimer, Seefeldt, and Dias 1999). Indeed, in comparison to the industrial strains, the
wild strain varieties of lactoccus and lactobacillus are more dependent on their own enzymatic
amino acid activity to survive, consequently their capacity to synthesize their own amino acids
is reflected on the amount of flavor compounds (such as VSC’s) that can be found in those
cheeses manufactured with them in parallel with the starter culture. Therefore, it is possible to
base this analysis on the fact that starter culture, adjunct bacteria and NSLAB all contribute to
the formation of methanethiol (Forde and Fitzgerald 2000; M. El Soda, Madkor, and Tong
2000; Seefeldt and Weimer 2000). Actually, the results of this work confirm that, and it can be
seen in figure 45 that the samples made with adjunct culture have significant higher
concentration of methanethiol. This can be explained by the many metabolic pathways that
adjunct culture can use to metabolize methionine, from which the cystathionine is the
principal one for most of the strains (Ayad et al. 1999; Seefeldt and Weimer 2000).
Nonetheless, the synthesis of MeSH via methionine γ-lyase is more efficient and is
characteristic of B. Linens. As a matter of fact it has been reported that lactococci can grow in
absence of cysteine but not methionine, and lactobacilli could not grow in the absence of
either, which indicate that both types of bacteria are auxothropic for both amino acids, but
their growth requirements are strain specific (Chopin 1993; Seefeldt and Weimer 2000).
Another fact that reinforces the last hypothesis is that these adjunct bacteria have potent
proteolytic systems composed by extracellular proteinase, endopeptidases, exopeptidases and
amino peptidases that increase proteolysis and remove amino acids from the amino terminal
ends of various peptides, which result in the increment of free amino acids such as methionine
(that was showed and explained in last section), which in high concentrations inhibits
cystathionine-lyase activity of Lc lactis spp. cremoris because the enzymes responsible for this
activity are biosynthetic and methionine inhibit their expression (Dias and Weimer 1998).
Therefore, it is possible to state that in cheese made with adjunct culture, methanethiol is
mainly provided by other strains of Lactococci, less susceptible to methionine concentration in
the growth medium, or by Lactobacilli, and/or B. Linens, which methionine presence had little
or no effect on cystathionine-lyase or methionine-γ-lyase activity.
Essentially the same principle describe above can be used to explain the higher levels of H2S
for the samples containing adjunct culture in figure 44. In addition to the H2S produced from
141
the sulfhydryl groups from thermal breakdown of cysteine, consequence of the denaturation of
whey protein that coagulate with caseins after thermal treatment of cheese milk, more H 2S is
generated when methionine is produce from cysteine through the β-elimination reaction
(Dobric et al. 2000; María Fernández et al. 2000), which is a reaction reproducible by other
lactococcus strains and certain genetic variants of lactobacillus helveticus, one of the possible
adjunct cultures (Smit, Smit, and Engels 2005; Lee et al. 2007).
Regarding DMS, it is way more difficult to explain the reason why there is and evident
increment in its concentration for the samples containing adjunct culture since its generation is
still poorly understood and unclear even for the starter culture strains.
Figure 41 VSC chromatogram for Cheese with adjunct culture
Figure 43 Development of MeSH in Cheese
made with and without adjunct culture
Figure 42 Development of H2S in Cheese made with and
without adjunct culture
142
Figure 44 Development of DMS in Cheese made with and
without adjunct culture
Effect of treatment on Proteolysis
The results of this work indicate that the methodology for monitoring proteolysis and
classifying Cheddar cheeses according to maturity and treatment can be based on
measurements of Total Kjeldahl Nitrogen (TKN) and a principal component analysis of the
RP-HPLC peptide profile of the WSN fraction. Contrary the Urea PAGE results do not
present evidence of differences in order to discriminate sample.
Soluble Nitrogen Fractions and TKN
Based on the same fractionation scheme describe in the last two chapters (Sousa, Ardö, and
McSweeney 2001; Voigt et al. 2012), it can be seen that nitrogen concentrations increased
during ripening time, and was different regarding the content of adjunct culture.
Figures 45 and 46 show that for all samples the nitrogen levels for the WSN fraction was the
highest in comparison to the TCA-SN and PTA-SN fractions. Similar to the cases of the
results in the last 2 chapters, the values collected were about 4 and 8 times higher respectively.
And after an observation period of 10 months, nitrogen concentrations were at least the double
of those by the beginning of the assessment
Regarding the TKN levels of WSN, in the figure 47 is possible to appreciate that during the
second and fourth month, the levels are higher for the samples made with adjunct culture, but
by the sixth month the difference became smaller and was sustained during the next 4 months.
143
However, it is clear that the samples containing adjunct culture have the tendency of
concentrating more nitrogen, and despite this fraction has been recognized as an index of
primary proteolysis (Bansal, Piraino, and McSweeney 2009), it is evident the influence of
NSLAB or adjunct culture enzymes in primary proteolysis, and although it is expected to be
expressed later in the ripening stage when numbers of LAB decrease and the adjunct culture
become dominant, its role in early proteolysis can be seen. However it is important to keep in
mind the TKN of the WSN fraction is a percentage of total Nitrogen and has no specific
information about the composition of the WSN fraction; therefore, these result could be a
synergic effect from different breakdown products and proteolytic agents. Nonetheless, the
results are in agreement with the works of Habibi-Najafi, Lee, and Law 1996; Laan et al.
1998; El-Soda, Madkor, and Tong 2000; Madkor, Tong, and El Soda 2000; that report higher
intracellular activity after de addition of lactobacilli strains and higher concentration of water
soluble nitrogen.
In figure 48, it can be seen that the levels of TCA-SN indicate differences in the proteolysis
development between samples containing adjunct cultures and the ones that do not.
Additionally, this figure show as well that the nitrogen levels in this fraction after 10 months
are about twice those by 2 months. So it is possible to appreciate how the difference in the
nitrogen extracted was larger as the extent of proteolysis was higher.
This fraction is a selective precipitation by TCA to fractionate peptides in the WSN. The
peptide solubility is related to hydrophobicity (Yvon, Chabanet, and Pélissier 1989), therefore,
it is expected that this fraction is rich in medium sized and small peptides, amino acids, which
can have low and medium hydrophobicity (T.K. Singh et al. 1994). Thus, it is possible that
many peptides derived from the N-terminal half of αs1-casein and the N-terminal half of βcasein might be extracted in this fraction; nonetheless, due to it is not known what was the
nature of the adjunct culture used, it is difficult to suggest a relation between the role of the
bacteria with the extra amount of nitrogen extracted, because it can be attributed to specificity
of the adjunct culture or to the rate of cell autolysis. In addition it might be possible that the
adjunct culture can be a combination of two NSLAB cultures that for sure will increase the
rate of proteolysis. Another interesting observation is that during the first 4 months the
144
amounts of total nitrogen are similar, which can be evidence of competition for substrate
between the starter and adjunct culture enzymes.
Regarding the PTA-SN fraction, it is possible to see that the amount of nitrogen extracted is
directly related to the extension of the ripening of samples, which increases along with time.
Also it can be seen an increment of 3 and 4 times the concentration at 2 months by the end of
the observation period (10 months) for the samples without and with adjunct culture
respectively. This observation is in agreement with the work of (C. M. Lynch et al. 1996),
which reports that the presence of lactobacilli led to increase the levels of small peptides and
amino acids, which is basically the composition of the PTA fraction, very small peptides
(<15 kDa) and amino acids of approximately 600 Da (Aston and Dulley 1982). However, as
in the case of the WSN fraction, this fraction is related to the extent of secondary proteolysis
and has no specific information about the composition. Nonetheless, by focusing the attention
on the first 4 months, it might be supported the observation in the TCA fraction about the
competition for the available substrate between starter and adjunct culture enzymes. Alike, the
TCA and WSN fraction it is really difficult to establish any relation between the results
obtained and the specific role of the adjunct culture without knowing its nature, because as it
is mentioned above, difference can be attributed to viable or to non-viable cells, or to the
combination of both.
From the results of these fractions, it is reasonable to affirm that it is possible to establish a
fair comparison between samples made with and without adjunct culture by using the
proposed fractionation scheme.
Figure 45 TKN fractions cheese with adjunct culture
145
Figure 46 TKN fractions cheese without adjunct culture
Figure 47 WSN Cheese with Adj culture Vs Cheese without
Figure 48 TCA-SN Cheese with Adj culture Vs Cheese without
146
Figure 49 PTA-SN Cheese with Adj culture Vs Cheese without
Peptide analysis by RP-HPLC
Based on the same procedure mentioned in the last 2 chapters for analyzing the raw data from
the peptide profile obtained by RP-HPLC by employing a PCA (Smith and Nakai 1990;
Benfeldt and Sørensen 2001; Piraino, Parente, and McSweeney 2004), it was possible to
establish a fair assessment of the proteolysis of Cheddar cheese samples prepared with and
without the addition of adjunct culture with a randomize design based in duplicate
observations.
The PCA results lead to a model with three principal components (PCs) that explain the 83.4%
of the variability of the data. The only combination that exposed correlation is the one for PC1
Vs PC2, which explain the 74.9% of the variability of the data and are related to the
discrimination of samples according age and adjunct culture treatment respectively as it can be
seen in figures 53 and 54. This indicate that the higher scores in PC1 correspond to samples
with longer ripening time while the higher scores in PC2 correspond to samples made with
addition of secondary culture, which means that proteolysis is faster for this cheeses. In
addition, the loading plot in the figure 55 shows high score for the segments 12-20, 20-25, 3540, 40-45 and 60-65 with regarding to PC1, which suggest that the relative amount of peptides
and amino acids eluting in this zones increase during ripening, whereas the scores for the
segments 25-30, 30-35, 45-50,50-55 and 55-60 indicate that the amount of peptides and amino
acids in these segments reduces over time. Additionally the scores for PC2 point out a increase
147
in the material eluting in the segments 12-20, 20-25, 35-40, 40-45, 45-50 and 50-55, and
reveal a reduction for the segments 25-30, 30-35, 55-60 and 60-65 as consequence of the
adjunct culture treatment.
Regarding intervals 12-20 and 20-25, the results in the loading plot indicate that the amount of
peptides and amino acids eluting in these segments increased as consequence of the
maturation and due to the addition of adjunct culture. This interpretation can be explain based
on the fact that proteolytic activity of secondary cultures complement the activity of the
starter, producing peptides with similar molecular weight and free amino acids (C.N. Lane and
Fox 1996a; C. M. Lynch et al. 1996; C. M. Lynch et al. 1999). Normally, these segments are
mostly composed by amino acids and small hydrophilic peptides product from the hydrolysis
of αs1-CN and κ-caseins followed by the action of the cell envelope proteinase (CEP) and
amino peptidase from LAB, resulting in peptides such as αs1-CN(f1-9) and αs1-CN(f1-13)
that accumulate during ripening, and others like αs1-CN (f1-8), αs1-CN(f8-23), αs1-CN(f923), αs1-CN(f14-23), αs1-CN(f10-?), αs1-CN(f17-?), αs1-CN(f18-?) and αs1-CN(f11-?) that
are N-terminal fragments from αs1-CN. Also, amino acids such as glutamic acid, valine,
isoleucine, leucine, lysine and proline can be found in this zone, which reveal the activity of
amino peptidases (Pep A), (Pep N) and proline iminopeptidase (T.K. Singh et al. 1994;
Manuela Fernández, Singh, and Fox 1998; Andersen, Ardö, and Bredie 2010). Therefore, it is
possible to suggest that the increment could be related to adjunct culture of lactoballi, which
have 5 to 100 times higher intracellular enzyme activity (Habibi-Najafi, Lee, and Law 1996;
Laan et al. 1998). In addition if the strain used has high peptidolytic potential with low
acidification ability, high levels of proteolysis can be expected (El-Soda, Madkor, and Tong
2000).
Contrary, the segments 25-30 and 30-35 got negative loadings for the PC1 and PC2, which
means that peptides eluting in these zones decrease over time and that the addition of the
secondary culture, apparently decreases their rate of proteolysis. This observation has to do
with the breakdown of αs1-CN, αs2-CN, αs1-CN (f24-199) and β-CN peptides (Tove M. I. E.
Christensen, Kristiansen, and Madsen 1989; Tanoj K. Singh, Fox, and Healy 1995; Tanoj K.
Singh, Fox, and Healy 1997; Benfeldt et al. 1997), which are related to enzymatic activity that
is essentially provided by Chymosin and Plasmin and has not so much relation to the
148
intracellular activity of adjunct culture. Therefore, it might be possible to say that the relative
decrease in the amount of eluents in these sections due to the adjunct culture can be related to
a general increase in the total amount of peptides. Additionally, in spite of the feasible early
accumulation of peptides such as αs1-CN (f85-91), αs1-CN (f11-?), αs2-CN (f170-?) and αs1CN (f175-182) resulting from the action of CEP and activity of endopeptidases (Pep O, Pep
F), it might be possible that the decreased amount of peptides in the 30-35 segment is related
to faster depletion of available substrate as consequence of the competition between the
proteolytic systems of LAB and the adjunct culture .
The intervals 35-40 and 40-45 exhibit fairly positive loading scores in relation to the PC1, and
slightly positive scores regarding PC2. This suggests that the relative amount of peptides in
these fractions rise with the cheese age and slightly increased due to the addition of adjunct
culture. Once again based on previous studies, these hydrophobic peptides could be: 1) αs1CN(f93–?), αs1-CN(f24–30), αs1-CN(f26–32), αs1-CN(f26–34) resulting from the hydrolysis
of the peptide αs1-CN(f24–199) by Chymosin, CEP and aminopeptidase in the segment; 2)
αs2-CN(f204–207), which is a C-terminal residue and product of lactooccal CEP (Fox et al.,
1994); and 3) β-CN(f45-52), as well reported as a product of hydrolysis by CEP and
aminopeptidase, and traces of γ-caseins (Tove M. I. E. Christensen, Kristiansen, and Madsen
1989; T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1995; Tanoj K. Singh, Fox, and
Healy 1997; Manuela Fernández, Singh, and Fox 1998). In addition, some amounts of amino
acids related to bitterness such as phenylalanine and histidine might be present in the last part
of the last segment (Manuela Fernández, Singh, and Fox 1998; Andersen, Ardö, and Bredie
2010). This makes sense since it has been reported that peptides eluting in this region of the
chromatogram correspond to high molecular mass molecules or molecules that contain
aromatic amino acids, which are characterized for being very hydrophobic (Gripon et al. 1991;
Gomez et al. 1997), and for eluting late in the peptide profiling by RP-HPLC and gel filtration
of the water soluble extract (Allan J. Cliffe, Marks, and Mulholland 1993). Therefore, a minor
increase in levels of proteolysis for these fractions has to do probably with the complementary
and supplementary enzymatic activity of adjunct culture, which might provide
homofermentative mesophilic lactobacilli that cause more bitterness (Habibi-Najafi, Lee, and
Law 1996) and consequently increased these values. Also some adjunct strains have high
149
peptidolytic activity (El-Soda, Madkor, and Tong 2000). However, it is important to keep in
mind that the role of these organisms in amino acid production requires more study.
Moreover, something really interesting is the reinforced reducing effect of material eluting in
the intervals 45-50 and 50-55 visible as negative scores for PC1 and positive for PC2, which
indicates that the relative amount of peptides and amino acids in this part of the chromatogram
reduces over time and is intensified by the addition of adjunct culture. As it was mentioned
above, material eluting in the last section of the chromatogram is related to bitter fractions
(Gripon et al. 1991; Allan J. Cliffe, Marks, and Mulholland 1993; Gomez et al. 1997), and in
addition it was reported by Singh et al. 1994; Manuela Fernández, Singh, and Fox 1998, that
the last fraction of the chromatogram mainly contained tryptophan, another aromatic amino
acid related to bitterness in cheese. Therefore this behavior can by basically explained based
on the findings of Habibi-Najafi, Lee, and Law 1996 and El-Soda, Madkor, and Tong 2000,
which points out the debitterase action and high potential to degrade hydrophobic aminoacids
and reducing bitter off-flavor of adjunct cultures, specially strains of L Helveticus, and mixed
strains of lactococci and Br. linens.
Finally the low scores in the remaining segments respect to PC1 and practically close to 0 for
PC 2, indicate that enzymatic activity associated to the reduction of the material eluting in this
zone, is apparently unaffected or less affected by the addition of adjunct culture. Indeed, the
enzymatic activity in this zone might be related to the catabolism of free amino acids.
Nonetheless because of the uncertainty associated to the type of adjunct culture employed in
the manufacture of the samples, it is really difficult to establish any accurate association.
However, it is important to recall that it is a common industrial practice to use isolated
cultures from good quality cheese where the criteria for selection is still vague, resulting in
potentially non reproducible results, which reminds the importance and need to identify the
proteolytic and lipolytic systems to be used. .
150
Figure 50 Scree plot of cheese with and without adjunct culture
Figure 51 Score plot of cheese with and without adjunct culture (age)
151
Figure 52 Score plot of cheese with and without adjunct culture (culture)
Figure 53 Loading plot of cheese with and without adjunct culture
152
Figure 54 Peptide profile cheese with adjunct culture (A and B) Vs Not (C and D) cheese, for 2 and
10 months
A
B
C
D
Electrophoresis
From the results of the Urea-PAGE it is possible to see that αs1-casein and β-casein are
breaking down during ripening. Indeed it can be seen that the degradation of αs1-casein is
faster and stronger than that of β-casein. However the primary proteolysis pattern for the
different observation is similar for αs1-casein and β-casein for the cheeses made with and
without adjunct culture. In addition, in the figures 57 to 59 it is possible to see a parallel
increase of those bands corresponding to γ-caseins over time. Therefore it is possible to say
that proteolysis patterns found are related to the activity of Chymosin and Plasmin rather than
associated to the activity of the starter or the adjunct culture
153
Figure 55 Urea PAGE for ripening of cheese made with and without adjunct culture
β -CN(f106-209)
β -CN(f29-209)
β -CN(f108-209)
β -CN
αs1 -CN
αz1-CN(f24-199)
αs1-CN(f121-199)
2M-A
2M-NA
4M-A
4M-NA
6M-A
6M-NA
154
Figure 56 Urea PAGE for ripening of cheese made with and without adjunct culture
6M-A
6M-NA
8M-A
8M-NA
10M-A
10M-NA
CONCLUSION
The present study demonstrates that the use of FFA profile, VSC’s profile, measurement of
the level of the Total Kjeldahl nitrogen for the WSN, TCA-SN and PTA-SN fractions, and the
PCA of the RP-HPLC peptide profile of the WSN fraction are effective tools and ripening
indices to differentiate Cheddar cheese samples regarding to their age and the adjunct culture
treatment. The urea-PAGE was effective to differentiate samples by their age; nonetheless it
is clear that it is not sensitive enough to detect differences related to the addition of adjunct
culture. On the other hand, , the results of levels of nitrogen for all the 3 fractions analyzed
demonstrated that proteolysis is faster for cheeses made with adjunct culture. This was
supported by the PCA model obtained which suggests differences caused by the role of the
secondary culture as supplement to the starter culture during ripening. Lipolysis was slower
for cheese produced without adjunct culture, which showed lower levels of individual FFA.
155
The amounts of DMS, H2S and MeSH showed differences between treatments and the
tendency to accelerate the catabolism of sulfur containing amino acids by including adjunct
culture. Once again, the results for DMDS and DMTS suggest that they are artifacts from
extraction and separation procedures rather than metabolites from the ripening of Cheddar
cheese.
REFERENCES
Andersen, Lene T., Ylva Ardö, and Wender L.P. Bredie. 2010. “Study of Taste-active
Compounds in the Water-soluble Extract of Mature Cheddar Cheese.” International
Dairy Journal V20 (8): pp. 528–536.
Aston, J. W., and John R. Dulley. 1982. “Cheddar cheese flavor review.” Australian Journal
of Dairy Technology V37 (2): pp. 59–64.
Ayad, Eman H.E, Annette Verheul, Catrienus de Jong, Jan T.M Wouters, and Gerrit Smit.
1999. “Flavour Forming Abilities and Amino Acid Requirements of Lactococcus
Lactis Strains Isolated from Artisanal and Non-dairy Origin.” International Dairy
Journal V9 (10): pp. 725–735.
Bansal, N, Paolo Piraino, and P. L. H. McSweeney. 2009. “Determination of Proteolysis in
Cheese.” In Handbook of Dairy Foods Analysis, pp. 405–417. CRC Press.
Benfeldt, Connie, John Sørensen, Katrine H. Ellegård, and Torben E. Petersen. 1997. “Heat
Treatment of Cheese Milk: Effect on Plasmin Activity and Proteolysis During Cheese
Ripening.” International Dairy Journal 7 (11): 723–731..
Benfeldt, Connie, and John Sørensen. 2001. “Heat Treatment of Cheese Milk: Effect on
Proteolysis During Cheese Ripening.” International Dairy Journal V11 (4–7): pp.
567–574.
Burbank, Helen M., and Michael C. Qian. 2005. “Volatile Sulfur Compounds in Cheddar
Cheese Determined by Headspace Solid-phase Microextraction and Gas
Chromatograph-pulsed Flame Photometric Detection.” Journal of Chromatography A
V1066 (1–2): pp. 149–157.
Chopin, Alain. 1993. “Organization and Regulation of Genes for Amino Acid Biosynthesis in
Lactic Acid Bacteria.” FEMS Microbiology Reviews V12 (1–3): pp. 21–37.
Christensen, Tove M. I. E., Kristian R. Kristiansen, and Jesper S. Madsen. 1989. “Proteolysis
in Cheese Investigated by High Performance Liquid Chromatography.” Journal of
Dairy Research V56 (5): pp. 823–828.
Cliffe, Allan J., Jonathan D. Marks, and Francis Mulholland. 1993. “Isolation and
Characterization of Non-volatile Flavours from Cheese: Peptide Profile of Flavour
Fractions from Cheddar Cheese, Determined by Reverse-phase High-performance
Liquid Chromatography.” International Dairy Journal V3 (4–6): pp 379–387.
Dias, Benjamin, and Bart Weimer. 1998. “Conversion of Methionine to Thiols by Lactococci,
Lactobacilli, and Brevibacteria.” Applied and Environmental Microbiology V64 (9):
pp. 3320–3326.
Dobric, Nada, Gäetan K.Y. Limsowtin, Alan J. Hillier, Nicholas P.B. Dudman, and Barrie E.
Davidson. 2000. “Identification and Characterization of a Cystathionine Β/γ-lyase
from Lactococcus Lactis Ssp. Cremoris MG13631.” FEMS Microbiology Letters
V182 (2): pp. 249–254.
156
El Soda, Morsi A. 1993. “The Role of Lactic Acid Bacteria in Accelerated Cheese Ripening.”
FEMS Microbiology Reviews V12 (1–3) (September): pp. 239–251.
El Soda, M., S.A. Madkor, and P.S. Tong. 2000. “Adjunct Cultures: Recent Developments
and Potential Significance to the Cheese Industry.” Journal of Dairy Science V83 (4):
pp. 609–619.
El-Soda, M., S. A. Madkor, and P. S. Tong. 2000. “Evaluation of commercial adjuncts for use
in cheese ripening: 4. Comparison between attenuated and not attenuated lactobacilli.”
Milchwissenschaft V55 (5): pp. 260–263.
Fernández, Manuela, Tanoj K. Singh, and Patrick F. Fox. 1998. “Isolation and Identification
of Peptides from the Diafiltration Permeate of the Water-Soluble Fraction of Cheddar
Cheese.” Journal of Agricultural and Food Chemistry V46 (11): pp. 4512–4517.
Fernández, María, Wim van Doesburg, Ger A. M. Rutten, Joey D. Marugg, Arno C. Alting,
Richard van Kranenburg, and Oscar P. Kuipers. 2000. “Molecular and Functional
Analyses of themetC Gene of Lactococcus Lactis, Encoding Cystathionine β-Lyase.”
Applied and Environmental Microbiology V66 (1): pp. 42–48.
Ferreira, A.D., and B.C. Viljoen. 2003. “Yeasts as Adjunct Starters in Matured Cheddar
Cheese.” International Journal of Food Microbiology V86 (1-2): pp. 131–140.
Forde, Amanda, and Gerald F Fitzgerald. 2000. “Biotechnological Approaches to the
Understanding and Improvement of Mature Cheese Flavour.” Current Opinion in
Biotechnology V11 (5): pp. 484–489.
Fox, P.F. 1989. “Proteolysis During Cheese Manufacture and Ripening.” Journal of Dairy
Science V72 (6): pp. 1379–1400.
Fox, P.F., and L. Stepaniak. 1993. “Enzymes in Cheese Technology.” International Dairy
Journal V3 (4–6): pp. 509–530.
Fox, P. F., J. M. Wallace, S. Morgan, C. M. Lynch, E. J. Niland, and J. Tobin. 1996.
“Acceleration of Cheese Ripening.” Antonie Van Leeuwenhoek 70 (2): 271–297.
Fox, PF, PLH McSweeney, and CM Lynch. 1998. “Significance of Non-starter Lactic Acid
Bacteria in Cheddar Cheese.” Australian Journal of Dairy Technology V53 (2): pp. 7–
8.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Gomez, Maria J., Sonia Garde, Pilar Gaya, Margarita Medina, and Manuel Nuñez. 1997.
“Relationship Between Level of Hydrophobic Peptides and Bitterness in Cheese Made
from Pasteurized and Raw Milk.” Journal of Dairy Research V64 (02): pp. 289–297.
Gripon, Jean-Claude, V Monnet, G Lambert, and M. J. Desmazeaud. 1991. “Microbial
enzymes in cheese ripening.” In Food enzymology, vol 1.
Habibi‐Najafi, Mohammad B., Byong H. Lee, and Barry Law. 1996. “Bitterness in Cheese: A
Review.” Critical Reviews in Food Science and Nutrition V36 (5): pp. 397–411.
Laan, Harry, Saw Eng Tan, Paul Bruinenberg, Gaëtan Limsowtin, and Malcolm Broome.
1998. “Aminopeptidase Activities of Starter and Non-Starter Lactic Acid Bacteria
Under Simulated Cheddar Cheese Ripening Conditions.” International Dairy Journal
V8 (4): pp. 267–274.
Lane, C.N., and P.F. Fox. 1996a. “Contribution of Starter and Adjunct Lactobacilli to
Proteolysis in Cheddar Cheese During Ripening.” International Dairy Journal V6 (7):
pp. 715–728.
Lee, Won-Jae, Dattatreya S. Banavara, Joanne E. Hughes, Jason K. Christiansen, James L.
Steele, Jeffery R. Broadbent, and Scott A. Rankin. 2007. “Role of Cystathionine βLyase in Catabolism of Amino Acids to Sulfur Volatiles by Genetic Variants of
157
Lactobacillus Helveticus CNRZ 32.” Applied and Environmental Microbiology V73
(9): pp. 3034–3039.
Lynch, C.M., P.L.H. McSweeney, P.F. Fox, T.M. Cogan, and F.D. Drinan. 1996.
“Manufacture of Cheddar Cheese with and Without Adjunct Lactobacilli Under
Controlled Microbiological Conditions.” International Dairy Journal V6 (8–9): pp.
851–867.
Lynch, C.M., D.D. Muir, J.M. Banks, P.L.H. McSweeney, and P.F. Fox. 1999. “Influence of
Adjunct Cultures of Lactobacillus Paracasei Ssp. Paracasei or Lactobacillus Plantarum
on Cheddar Cheese Ripening.” Journal of Dairy Science V82 (8): pp. 1618–1628.
Madkor, S.A., P.S. Tong, and M. El Soda. 2000. “Ripening of Cheddar Cheese with Added
Attenuated Adjunct Cultures of Lactobacilli.” Journal of Dairy Science V83 (8): pp.
1684–1691.
Manning, D. J., Helen R. Chapman, and Zena D. Hosking. 1976. “The Production of Sulphur
Compounds in Cheddar Cheese and Their Significance in Flavour Development.”
Journal of Dairy Research V43 (2): pp. 313–320.
McSweeney, P.L.H., P.F. Fox, J.A. Lucey, K.N. Jordan, and T.M. Cogan. 1993. “Contribution
of the Indigenous Microflora to the Maturation of Cheddar Cheese.” International
Dairy Journal V3 (7): pp. 613–634.
McSweeney, Paul L.H., and Maria José Sousa. 2000. “Biochemical Pathways for the
Production of Flavour Compounds in Cheeses During Ripening: A Review.” Le Lait
V80 (3): pp. 293–324.
Piraino, Paolo, Eugenio Parente, and Paul L. H. McSweeney. 2004. “Processing of
Chromatographic Data for Chemometric Analysis of Peptide Profiles from Cheese
Extracts:  A Novel Approach.” Journal of Agricultural and Food Chemistry 52 (23):
6904–6911.
Seefeldt, K.E., and B.C. Weimer. 2000. “Diversity of Sulfur Compound Production in Lactic
Acid Bacteria.” Journal of Dairy Science V83 (12): pp. 2740–2746.
Singh, T.K., P.F. Fox, P. Højrup, and A. Healy. 1994. “A Scheme for the Fractionation of
Cheese Nitrogen and Identification of Principal Peptides.” International Dairy
Journal V4 (2): pp. 111–122.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1995. “Water-Soluble Peptides in Cheddar
Cheese: Isolation and Identification of Peptides in the Diafiltration Retentate of the
Water-Soluble Fraction.” Journal of Dairy Research V62 (4): pp. 629–640.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1997. “Isolation and Identification of Further
Peptides in the Diafiltration Retentate of the Water-soluble Fraction of Cheddar
Cheese.” Journal of Dairy Research V64 (3): pp. 433–443.
Smit, Gerrit, Bart A. Smit, and Wim J.M. Engels. 2005. “Flavour Formation by Lactic Acid
Bacteria and Biochemical Flavour Profiling of Cheese Products.” FEMS Microbiology
Reviews V29 (3): pp. 591–610.
Smith, A. M., and S. Nakai. 1990. “Classification of cheese varieties by multivariate analysis
of HPLC profiles.” Canadian Institute of Food Science and Technology Journal V23
(1): pp. 53–58.
Sousa, M.J, Y Ardö, and P.L.H McSweeney. 2001. “Advances in the Study of Proteolysis
During Cheese Ripening.” International Dairy Journal V11 (4-7): pp. 327–345.
Urbach, Gerda. 1993. “Relations Between Cheese Flavour and Chemical Composition.”
International Dairy Journal V3 (4–6): pp. 389–422.
Voigt, Daniela D., François Chevalier, John A. Donaghy, Margaret F. Patterson, Michael C.
Qian, and Alan L. Kelly. 2012. “Effect of High-pressure Treatment of Milk for
158
Cheese Manufacture on Proteolysis, Lipolysis, Texture and Functionality of Cheddar
Cheese During Ripening.” Innovative Food Science & Emerging Technologies V13
(0): pp. 23–30.
Wallace J.M., and Fox P.F. 1997. “Effect of Adding Free Amino Acids to Cheddar Cheese
Curd on Proteolysis, Flavour and Texture Development.” International Dairy Journal
V7 (2): pp. 157–167.
Weimer, Bart, Kimberly Seefeldt, and Benjamin Dias. 1999. “Sulfur Metabolism in Bacteria
Associated with Cheese.” Antonie Van Leeuwenhoek 76 (1): 247–261.
De Wit, M., G. Osthoff, B.C. Viljoen, and A. Hugo. 2005. “A Comparative Study of Lipolysis
and Proteolysis in Cheddar Cheese and Yeast-inoculated Cheddar Cheeses During
Ripening.” Enzyme and Microbial Technology V37 (6) (November 1): pp. 606–616.
Yvon, Mireille, Claire Chabanet, and Jean-Pierre Pélissier. 1989. “Solubility of Peptides in
Trichloroacetic Acid (TCA) Solutions Hypothesis on the Precipitation Mechanism.”
International Journal of Peptide and Protein Research V34 (3): pp. 166–176.
159
CHAPTER 5
COMPARATIVE CHEMICAL ANALYSIS OF GOOD AND WEAK
CHEDDAR CHEESE DURING RIPENING
Lemus Muñoz, Freddy Mauricio, Qian Michael
Department of Food Science and Technology, Oregon State University
Corvallis Oregon 97331
160
ABSTRACT
Cheddar cheese samples graded as good and/or weak by a trained panel were analyzed during
ripening. Proteolysis was studied by a fractionation scheme, resulting in an insoluble fraction
analyzed by urea polyacrylamide gel electrophoresis (Urea-PAGE), and a soluble fraction
which was further investigated through water soluble nitrogen (WSN), trichloroacetic acid
soluble nitrogen (TCA-SN) and phosphotungstic acid soluble nitrogen (PTA-SN) analyzed by
total Kjeldahl nitrogen content (TKN). Reversed phase high performance liquid
chromatography (RP-HPLC) was used to study the peptide profile of the water soluble
fraction. Lipolysis was studied by levels of individual free fatty acids determined through gas
chromatography-flame ionization detection (GC-FID) after isolation employing solid phase
extraction (SPE). Volatile sulfur compounds were studied using head space solid phase microextraction (SPME) coupled with gas chromatography-pulsed flame photometric detection
(PFPD).
It was found that Urea-PAGE is capable to differentiate samples according their age, but it
could not be used to discriminate samples regarding their quality. Nonetheless, measurements
of total Kjeldahl Nitrogen (TKN) of the WSN, TCA-SN, and PTA-SN fractions, and the
principal component analysis of the RP-HPLC peptide profile of the WSN fraction, revealed
differences in the rate and pattern of proteolysis for the samples. Good cheese, developed
higher level of total nitrogen for the WSN, TCA-SN and PTA-SN fractions, indicating that
primary and secondary proteolysis were faster for these samples during ripening. It was
obtained a PCA model with 3 principal components that accounted for the 80.7% of the
variability from data collected. This model discriminate the samples according age and
quality, suggesting the samples undergo more or faster proteolysis. In addition, FFA profiles
demonstrated higher levels of low and medium chain free fatty acids for good cheese, which
suggest faster lipolysis during ripening. The Volatile Sulfur Compounds (VSC) analysis
showed higher levels of DMS and MeSH and lower levels of H2S, suggesting slower
catabolism of sulfur containing amino acids in weak cheese.
161
INTRODUCTION
The production of Cheddar cheese with constant quality during the year is limited due to
natural variation of milk composition or the extent of the enzymatic activity during ripening.
However, in terms of homogeneity in order to satisfy customer expectation and to assure a
reliable production, it is necessary to document and understand the chemical composition
differences related to samples characterized as good or weak according to a experienced
sensory panel. As a matter of fact, the traditional scope to assess quality based on off-flavor
and defects (O’Shea, Uniacke-Lowe, and Fox 1996), could be complemented by monitoring
biochemical changes occurring during ripening, which also allow to identify variations in
manufacturing practices (M E Carunchia Whetstine et al. 2007; J. M. Lynch, Barbano, and
Fleming 2002) resulting in perceptible disparity of flavor and texture in the ready to sale
product. Indeed, currently most of the assessment of cheddar cheese quality is done by trained
sensory panels that expensive, time consuming and essentially depend on the presence or
absence of defects, leading to results that are subjective rather than objective. Thus, better
grading implies the use of accurate measurements and reliable instrumental methods to predict
and determine the flavor quality of cheese.
Cheddar cheese flavor is a balance of several volatile and non-volatile sapid compounds
(Engels et al. 1997; Curioni and Bosset 2002). The volatile fraction contributes to its aroma
and the water soluble fraction is responsible for its taste (Aston and Dulley 1982; Aston and
Creamer 1986). Thus sample differentiation based on proteolysis and lipolysis implies the
separation, characterization and quantification of peptides, amino acids, free fatty acids and
another key volatile compounds using chromatographic methods such as RP-HPLC, GC-MS,
GC-PFPD and GC-FID, electrophoretic methods, and other emerging technologies such as
Fourier transform infrared (FT-IR) spectroscopy (Smith and Nakai 1990; Dimos 1992;
Subramanian, Harper, and Rodriguez-Saona 2009).
It has been reported significant correlations between levels of pH 4.6-soluble nitrogen,
phosphotungstic acid (PTA)-soluble nitrogen and trichloroacetic acid (TCA)-soluble nitrogen
and the age, flavor intensity, and flavor development in Cheddar cheese (O’Shea, UniackeLowe, and Fox 1996; Are Hugo Pripp, Stepaniak, and Sørhaug 2000; Upadhyay et al. 2004).
In addition, RP-HPLC has been effectively used to identify cheese variety and to determine its
162
age because of its high resolution power, reproducibility and low time consumption (Bican &
Spahni, 1991).
Thus, in order to contribute to the understanding of the flavor development of Cheddar cheese
during ripening, the aim of this study was to use complementary approaches to objectively
evaluate and correlate samples of different quality and age by using analytical methods such
as GC-PFPD, GC-FID and HPLC, Urea-polyacrilamide gel electrophoresis and determination
of nitrogen content of different soluble fractions.
MATERIALS AND METHODS
CHEESE SAMPLES
Cheeses samples were manufactured by Tillamok county creamery according with
standard protocols. Blocks of cheese of different quality were randomly selected from
three consecutive manufacturing days. All samples are stored at (-37C) to stop ageing
process until analysis is completed.
FREE FATTY ACIDS ANALYSIS
Chemicals
Pentanoic acid, heptanoic acid, nonanoic acid, undecanoic acid, and heptadecanoic acid were
used as internal standards, they were purchased from Eastman (Rochester, N.Y., U.S.A).
Butanoic acid, hexanoic acid, octanoic acid, decanoic acid, dodecanoic acid, tetradecanoic
acid, 9-tetradecanoic acid, hexadecanoic acid, 9-hexadecanoic acid, octadecanoic acid, 9octadecanoic acid, 9,12-octadecanoic acid and 6,9,12 octadecanoic acid were used for the
standard stock solution, and were obtained from Aldrich Chemical Co. Inc (Milwaukee,
Wisconsin, U.S.A). Heptane, Isopropanol, Sulfuric acid, anhydrous sodium sulfate,
chloroform, formic acid and diethyl ether were obtained from Fisher.
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Six
grams of this previously freeze-ground cheese, 1 ml of 2N sulphuric acid and 1 ml of internal
163
standard solution (C5:0, C7:0, C9:0, C11:0 and C17:0 in 1:1 hetpane-isopropanol) were mixed
with 7 grams of anhydrous sodium sulfate and 20 ml of 1:1 diethyl ether- heptane in a 40 ml
amber vial using a sonicator and manual agitation. During sonication, the salt-slurry solution
is initially exposed for 15 minutes, after which each vial is shake vigorously to continue with a
second sonication period of 20 minutes. With a glass-Pasteur pipette, the sample extract
(solvent) is transferred to an AccuBOND amino cartridge (Agilent Technologies) conditioned
previously with 10 ml of heptane. After the addition of the sample, the column is washed with
5 ml of 2:1 Chloroform-Isopropanol to remove non volatile triglycerides and phospholipids
using a manifold vacuum chamber. Once the washing step is complete, free fatty acids are
eluted with 5ml of 2% formic acid in diethyl ether, collected in a 20 ml vial and stored in the
freezer until GC analysis.
Chromatography
The analysis was performed using a Hewlett-Packard 5890 gas chromatograph equipped with
a flame ionization detector (FID). Samples were analyzed on a DB-FFAP column (15m x
0.53mm ID, 1 m film thickness; Supelco Wax10, Supelco U.S.A). Injector and detector
temperatures were 250C. Nitrogen was used as carrier gas at a flow rate of 15 ml per minute at
a split ratio of 1 to 1. The oven temperature was programmed for a 2 minutes hold at 60C,
raised to 230C at a rate of 8C per minute with a hold of 20 minutes at 230C.
Quantitative analysis
The levels of free fatty acids concentrations were calculated based on individual peak area
from GC-FID response in comparison to the internal standard peak area, by using standard
calibration curve of individual free fatty acid using Peak Simple software (SRI instruments,
Torrance, CA). Each experimental value corresponds to the average of the 3 extraction
replicates.
VOLATILE SULFUR COMPOUNDS (VSC’S)
Chemicals
Dimethyl sulfide (DMS) was purchased from TCI America (Portland, OR, U.S.A.); gaseous
methanethiol (MeSH) was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wisconsin,
U.S.A), and a solution was prepared by bubbling the gas into cold methanol; a H2S solution
164
was prepared by dissolving Na2S.9 H2O (Sigma Co) in acidic water stabilized with citric acid
(pH 3).
Extraction
From each 2lb block of cheese, 100 grams were wrapped in alumina foil, frozen with liquid
nitrogen during 6 minutes, and then grinded for 30 seconds to obtain a fine powder. Then one
gram of this freshly prepared powder is added to a 20ml vial (formerly flushed with argon),
followed by the addition of 4 ml of 1M citric acid and 20 l of the internal standard solution.
After addition of sample vials were immediately sealed with screw caps with teflon-lined
silicone septa. The vials used in this study were previously deactivated with DMTCS 5%
solution in toluene, toluene, methanol and distillate water.
The volatile sulfur compounds were extracted with an 85 m Carbox-PDMS fiber (Supelco,
Bellefonte, PA, U.S.A.). Prior to use, the fiber was conditioned at 300 C for 90 minutes. The
fiber was then placed into a SPME adapter of a CombiPAL autosampler (CTC analytics AG,
Zwingen, Switzerland) Fitted with a vial heater/agitator. Samples were pre-equilibrated at 500
RPM at 40C for 5 minutes, and the extraction of VSC’s was done at 250 RPM at 40C for 25
minutes. The desorption time was 5 minutes and 30 seconds.
Chromatography
The analysis was performed using a Varian CP-3800 gas chromatograph (Varian, Walnut
Creek, CA, U.S.A.) equipped with a pulsed flame photometric detector (PFPD). The
separation of analytes was made using a DB-FFAP fused silica capillary column (30m, 0.32
mm ID and 1 m film thickness; Agilent, Palo Alto, CA, USA) and nitrogen as carrier gas at
constant flow at 2 ml per minute. The injector temperature was 300 C and it was in the
splitless mode. The oven temperature was programmed for a 3 minutes hold at 35C, raised to
150C at a rate of 10C per minute, held for 5 minute, and then heated to 220C at a rate of 20C
per minute with a final hold of 3 minutes. The PFPD was held at 300 C and 450 V with the
following flow rates: Air 1 at 17 ml per min, H2 at 14 ml per min, and Air 2 at 10 ml/min. The
detector response signal was integrated using the software Star Workstation 6.2, Varian)
165
Quantitative analysis
Matrix effect
In order to retain the matrix effect during the construction of the calibration curves, cheese
powder from the “youngest sample” is used. It is de-volatilized by exposure to room
conditions in a hood for 2 hours. Then 1 gram of powder is added to 4 ml of 1M citric acid in
a 20 ml vial and exposed to a 50C water bath for 30 mins, prior to the addition of standards
and internal standard solutions.
Sulfur standards and internal standard preparation
Two internal standards were used for the quantification of VSC’s: ethyl methyl sulfide (EMS)
for H2S, MeSH and DMS, and isopropyl disulfide (IsoProDS) for DMDS and DMTS. The
concentration of the internal standard solution was 500 ppm EMS and 500 ppm IsoProDS in
methanol. Calibration curves were constructed by spiking cheese samples with a range of
known concentrations of H2S, MeSH and DMS. Hydrogen sulfide (H2S) was prepared by
dissolving Na2S.9 H2O in acidic water (pH = 3). Different concentrations of sodium sulfide
solutions were made, and the concentrations of H2S were calculated based on the amounts of
salt added to the matrix. A standard solution of 100 ppm of DMS was individually prepared in
cooled methanol (-15C), and dilutions were made with cooled methanol at the same
temperature. The mesh standard was prepared as following: 1) newly deactivated, recently
flushed with argon, and cooled vials were used; 2) The original standard solution was made by
bubbling pure MeSH into cooled methanol; 3) Dilutions were prepared by taking aliquots
from the original solution contained in a sealed vial, through the teflon-lined silicone septa by
using a syringe. And then injecting the aliquots into new sealed vials containing proportional
amount of cooled methanol through the septa; 4) 1 gr of devolatilized cheese is added to a
recently flushed vial (argon was used), which is immediately flushed again; 5) simultaneous
argon flushing and addition of 4ml of “free” dissolved oxygen-1M citric acid solution and
quick sealing of the vials; 6) Addition of 20 l of internal standard and MeSH standard
through septa. The identification of target compounds was made by comparing retention times
with those of pure standards. Ratios of the square root of the standard area to the
corresponding square root of the internal standard area were plotted Vs concentration ratios to
166
determine the relation between the response and concentration for the unknowns. Triplicate
analysis was performed for all samples
PROTEOLYSIS
Chemicals
Sulfuric acid was purchased from Fisher Scientific International Inc. (Pittsburgh, PA, U.S.A.);
Trichloroacetic acid was purchased from Alfa Aesar (Royston, UK); and phosphotungstic acid
was obtained from Aldrich Chemical Co. Inc (Milwaukee, Wis, U.S.A)
Sample preparation and fractionation
From each 2lb block of cheese, 60 grams are blended with 120 ml of distilled water pre-heated
to 55C. The mixture is blended for 5 minutes and the homogenate is incubated at 55C for 1
hour. Then the pH is adjusted to 4.6 with 1M HCl and the mixture is centrifuged at 3000g for
30 minutes at 4C. Suspension and supernatant were filtered thoroughly 3 times through glass
wool. The filtrate was safe at -20C for macro blog digestion method analysis, and RP-HPLC
analysis. The insoluble pellet was frozen at -20C for further Urea-PAGE gel electrophoresis
analysis.
The trichloroacetic acid soluble nitrogen fraction (TCA-SN) was prepared by the addition of
25 ml of pH 4.6 soluble fraction (WSN) to 25 ml of 24% trichloroacetic acid solution. Then
the mixture is equilibrated for 2 hours at room temperature and filtered through filter paper
Whatman No 40 before macro blog digestion method analysis.
For the phosphotungstic acid soluble nitrogen fraction (PTA-SN), 10 ml of WSN are added to
7 ml of 3.95 M H2SO4 and 3 ml of 33% phosphotungstic acid solution. Then the mixture is
equilibrated overnight at 4C and filtered through filter paper Whatman No 40 before macro
blog digestion method analysis.
Duplicate analysis was performed for all samples.
167
Macro blog digestion (Kjeldahl Digestion)
From the fractions collected an aliquot (2 ml for the Water soluble fraction, 1ml for TCA-SN
and 1 ml for PTA-SN) is added into a 70 ml Kjeldahl Digestion flask with 10 ml of H2SO4 and
the catalyst pellet containing 0,075 and 1,5 grams of mercuric oxide and potassium sulfate
respectively. The mixture is warmed to 150 C and hold for 1 hour, then heated to 250 C and
hold for 1 hour, and finally heated to 350 C and hold for 2 hours. After digestion the sample is
cooled down overnight to room temperature, and diluted with distillate water to 70 ml,
followed by a gentile agitation. Then a 5 ml aliquot is used to determine the nitrogen content
by a rapid flow analyzer FOSS II.
Reversed phase High performance liquid chromatography analysis
The RP-HPLC analysis was performed using a Shimadzu 6 series liquid chromatograph
(Shimadzu scientific instruments, Kyoto Japan), consisting of an autosampler, 2 pumps, a
multi-wavelength spectrophotometer and a controller unit. It was used a nucleosil RP-8
analytical column (250x 4mm, 5 m particle size, 300 A pore size) and a guard column (4.6
x10 mm) from waters (Milford, MA, U.S.A.). The mobile phase consists of solvent A (0.1%
TFA in deionized and vacuum filtered water) and solvent B (0.1% TFA in acetonitrile). The
elution was monitored at 214nm. The following gradient elution was performed: 1) 100%
solvent A for 5 minutes followed by a linear gradient to 55% solvent B (v/v); 2) elution at
55% solvent B for 6 minutes followed by a linear gradient to 60%; 3) elution at 60% solvent B
for 3 minutes; 4) The column is washed using 95% solvent B during 5 minutes; 5) the column
is equilibrated using 100% solvent A during 10 minutes. The sample (WSN fraction) was
dissolved in solvent A (10 mg per ml) and then micro-centrifuged at 14000 RPM for 10
minutes. An aliquot of 40 l from the extract was injected to a flow rate of 0.75 ml per min.
Electrophoresis
Samples of the water-insoluble nitrogen fraction were dry frozen prior to analysis. Samples
were dissolved in a buffer (0.75 g tris, (hydroxymethyl) methylamine, 49 gr urea and 0.4 ml
concentrated HCl, 0.7 ml 2-mercaptoethanol and 0.15 gr bromophenol blue, dissolved to
100ml) and hold at 50C for 40 min. Urea-polyacrylamide gel electrophoresis (urea-PAGE)
was carried out using a Protean II xi cell vertical slab unit (Bio-Rad Laboratories ltd., Hemel
Hempstead, Herts, UK). Urea-PAGE gels (12.5%) were prepared and run according to the
168
method or Ardö (1999). Reagents used were obtained by Sigma-Aldrich, Inc and Fisher
Scientific.
STATISTICAL ANALYSIS
A two-way analysis of variance (ANOVA) on data was carried out using a general linear
model procedure with Turkey’s pair wise comparison at 95% confidence level, using the
package Minitab 15 (minitab Ltda., Coventry, UK).
RESULTS AND DICUSSION
FREE FATTY ACIDS FFA
The levels of lipolysis tracked by the amount of individual FFA showed differences between
the cheeses graded as good and weak. However, in both cases it can be seen the tendency of
FFA to increase during maturation of samples. Indeed, it is evident in figures 58, that short
chain FFA tend to increase faster, reaching concentrations about 4 times the initial one during
the observation period. This can be related to more mobility and better access of enzymes to
these substrates, which are essentially located at the positions sn-1 and sn-3 of the
triacylglicerides (Balcão and Malcata 1998). On the other hand, this behavior might suggestes
that enzymatic activity is most likely dominated by lipases since they are specific for the outer
ester bonds of tri or diacylglycerides (Deeth and Touch 2000a). In addition, in spite of the low
change, long chain FFA increased their concentration during ripening too, displaying as well
higher levels for good cheese. which seems related to lipolytic activity rather than estereolityc
activity.
Although the most dominant peaks in figure 57 correspond to C14:0, C16:0, C18:0 and
C18:1, due to their significant lower odor thresholds (Molimard and Spinnler 1996a), they are
not considered as important contributors to the overall aroma of Cheddar cheese. As a matter
of fact, this relative quantitative relevance has to do with the fact that these FFA are the most
abundant ones in raw milk (Yvonne F. Collins, McSweeney, and Wilkinson 2003). Contrary,
in the case of short chain FFA such as C4:0 and C6:0, or FFA such as C8:0, C10:0 and C12:0,
despite of their lower concentration, they contribute directly and indirectly to the characteristic
aroma of Cheddar cheese, and it was evident the trend of good cheeses to developed higher
169
levels, specially for the 6, 9 and 12 month of maturation. Their rate of generation mainly
depend on enzymatic activity; however, based on the fact that the most important lipolytic
activity is provided by LAB enzymes, composed by esterases and lipases, it is not possible to
tell which one had higher influence in the lipolysis of these samples without a study of
specificity. Nonetheless, in order to explain the difference between these 2 types of samples, it
might be a better approach to focus on variables that contribute to the decriment of the LAB
enzymatic activity such as temperature and relative humidity of ripening rooms, and
differences during salting (which for sure could have inhibitory effect since LAB enzymes are
really sensitive to the salt in moisture content (Gripon et al. 1991; P. F. Fox and Stepaniak
1993)), rather than look at other manufacturing operation such as the heat treatment or
standardization of milk which do not have a direct impact on intracellular enzymes. Another
factor related to the lower lipolysis of weak cheese can be differences in the cell viability and
autolysis of the starter strain, which indeed might suggests the use of different starter during
the acidification in the manufacturing of these samples.
Figure 57 FFA chromatogram week cheese
170
Figure 58 Development of individual FFA in good cheese vs weak cheese
171
172
Volatile Sulfur Compounds VSC’s
In addition to free fatty acids (FFA), Volatile sulfur compounds (VSC’s) correlate with good
Cheddar cheese flavor (B. Weimer, Seefeldt, and Dias 1999). When smelled alone they smell
like garlic, onion, cabbage and skunk, but when they are mixed, they contribute to pleasant
Cheddar cheese flavor notes. They result from decomposition of sulfur containing amino acids
such as cysteine and methionine. Therefore this is another biochemical event occurring during
Cheddar cheese ripening that can be use to track the extension of the maturation of samples.
Indeed, It has been reported that high concentrations of H2S, MeSH, and DMS are found in
Cheddar Cheese, while DMDS, DMTS and 3-methylthiopropionaldehyde have low
concentrations. Other compounds such as Carbonyl sulfide, carbon disulfide and dimethyl
sulfone are not important contributors (H. M. Burbank and Qian 2005).
Subsequent to a thorough de-activation of injection liner and vials to prevent methanethiol
(MeSH) oxidation, and an adecuate stabilization of analytes by using an organic acid buffer
solution (citric acid 1M), the results from this work suggest that only hydrogen sulphide
(H2S), carbon disulphide (CS2), MeSH, and dimethyl sulphide (DMS) were developed during
ripening, and only small and negligible amounts of dimethyl disulphide (DMDS) and dimethyl
trisulphide (DMTS) were scarcely found in the chromatograms for the samples analyzed.
Which is the reason why only the development of MeSH, DMS and H2S will be discussed.
173
Standard calibration curves were calculates for each compound, and it is possible to observe in
figures 60 and 61 that good linear correlation coefficients were obtained for H2S and DMS.
Nonetheless it was not possible to achieve a descent calibration curve for MeSH due to its
oxidation to DMDS and DMTS, therefore the interpretation of results for this last one was
based on the area ratio between MeSH and the internal standard EMS, instead of using its
concentration
The results in figures 62, 63 and 64 show a steady development for all the compounds and
samples excepting H2S good cheese, which apparently did not increase so much during the
observation period. Also, this figures demonstrate that there are considerable differences of
the sulfur attributes related to the quality of the samples graded by trained panel.
In figure 63, hydrogen sulfide did not show a steady development for the good samples.
Moreover , weak samples displayed a higher concentration of H2S. However, the difference
between these types of samples during the initial stage of the ageing process was absolutely
not significant, but after 9 months it became noticible and evident. Therefore, in the case of
good cheese it was difficult to establish any trend during the ripening in contrast to weak
cheese, which increases H2S concentration along the maturation process. In addition, based on
the fact that the H2S sensory threshold is 10 ppb in water (Rychlik et al. 1998) and the
concenntration for weak samples varied from 20 to 30 ppm it was possible to confirm its role
as key contributor to the cheddar cheese aroma. The higher concentration for weak samples
could be related to 1) differences in the cheese milk, either because of the standardization
process or due to the heat treatment, which potentially can incorporate β-lactoglobulins to the
casein micelles and consequently increase the amounts of cysteine, which along with
methionine are the main precursors of H2S (B. Weimer, Seefeldt, and Dias 1999; Lee et al.
2007; del Castillo-Lozano et al. 2008). Indeed, (Hutton and Patton 1952; K. R. Christensen
and Reineccius 1992) reported that the concentration of H2S in milk increases linearly with
heating temperature; and 2) changes related to the LAB and NSLAB microflora, which supply
enzymes such as methionine-γ-lyase, cystathionine-β-lyase and cystathionine-γ-lyase that
produce methanethiol and H2S.
174
Regarding MeSH, figure 62 indicates a significant difference between samples, showing
higher concentration for the good samples. Unfortunately, it was not possible to confirm
MeSH as a potent odorant in cheese due to the lack of effectiveness in constructing a
calibration curve free of its oxidation products DMDS and DMTS. Nonetheless, the analysis
based on area ratios suggests a steady concentration increase for both types of samples in
agreement with Urbach 1995. As in the case of H2S, the higher concentration for the good
samples can be related to more enzymatic activity of LAB and NSLAB, and to a possible
higher availabilty of substrate. This last possibility can be consequence of incorporation of
whey protein to the casein micelles during the heating treatment of cheese milk, or due to
adjustments to the protein content of the cheese milk during standardization, which from the
results seems like it is different for the manufacturing process of these to types of samples. On
the other hand, because it has been proposed that MeSH is enzymatically produce rather than
by chemical reactions (Alting et al. 1995; Smacchi and Gobbetti 1998; Dias and Weimer
1998), it might be pertinent and more likely to attribute the different response to a raise in the
L-methionine γ-lyase and/or cystathionine β-lyase and γ-lyase activity (Alting et al. 1995)
(Tanaka, Esaki, and Soda 1985) resulting from, 1) a milder heating treatment of the cheese
milk, which results in a higher population of indigenous bacteria, 2) addition of adjuct
cultures, or 3) the use of LAB starters with different specificity and autolysis ability.
In a similar way to MeSH, figure 64 shows a steady increment of the concentration of DMS
for both types of samples during maturation. Aslo, it can be observed a noticeable difference
between the good and weak samples,
with higher concentration for the first type.
Additionally, it can be seen that this sulfur compound is a key contributor to the overall flavor
of Cheddar cheese since the concentrations found range between 10 and 45 ppm, amounts
comparable to the ones obtained by Burbank and Qian (2008), while the sensory threshold is
2ppm in water (Rychlik et al. 1998).
The higher DMS concentration for good samples is difficult to explain since its generation
mechanism has not been well understood yet. And in spite that it is known that DMS
concentration in raw milk is significant and is influenced by the diet of the cows (Manning et
al. 1976; Forss 1979), and it can be generated from sulfhydryl group of milk proteins, mainly
β-lactoglobulin and if present the milk fat globule membrane proteins, where methionine is
175
most likely the precursor for its generation after protein thermal denaturation (Datta et al.
2002), it is not possible to suggest other reasons why good cheese has higher concentration
more than: 1) a completely different type of cheese milk, which could be supported by the fact
that the initial observation point displays a obvios difference in comparision to those from the
H2S and MeSH graphs; and 2) elevated numbers of secondary micro flora such as propionibacteria, present in the milk microflora (Baer and Ryba 1992), resulting from different
approaches to the heating treatment of milk, leading to more formation of DMS from
methionine (Curioni and Bosset 2002); or other bacteria sucha s Brevibacterium linens (Dias
and Weimer 1998), different strains of Lactococcus lactis subsp. cremoris (Alting et al. 1995),
Lactococcus lactis subsp. lactis, Lacto bacillus sp., Propionibacterium shermanii and/or the
yeast Geotrichum candidum and Kluyveromyces lactis (K. Arfi et al. 2002) that could as well
to provide enzymes such as thiol transferases to promote the conversion of some MeSH into
DMS (Bentley and Chasteen 2004).
Regarding the general absence of the sub-products DMDS and DMTS in the results of this
study, it might be possible to say that based on the thorough sample preparation work and the
fact that cheese has a low redox potential, -150 to -200 mV, (Donald J. Manning and Moore
1979; Green and Manning 1982), these compounds might not be generated during the ripening
of curds, and instead their formation is the simple consequence of oxidation of MeSH once the
cheese is exposed to a non controlled environment.
Figure 59 VSC's chromatogram Good Cheese
176
Figure 60 Calibration Curve H2S
Figure 61 Calibration Curve DMS
Figure 62 Development of MeSH in Good and
weak cheese
177
Figure 63 Development of H2S in Good and weak
cheese
Figure 64 Development of DMS in Good and weak cheese
Effect of treatment on Proteolysis
Cheddar cheese graded as “Good” and “Weak” by a trained panel was investigated during a 15
month maturation period through: 1) measurements of Total Kjeldahl Nitrogen (TKN) of the
water soluble nitrogen (WSN), trichloroacetic acid soluble nitrogen (TCA-SN) and
phosphotungstic acid soluble nitrogen (PTA-SN) fractions; 2) RP-HPLC peptide profiles of
the WSN fraction;
and 3) Urea-PAGE peptide profiles of the water insoluble nitrogen
fraction. The results reveal clear differences in the rate and pattern of proteolysis. Nonetheless,
from the three methods employed to evaluate the extend of proteolysis, the TKN of fractions
178
WSN, TCA-SN and PTA-SN, and the peptide profile analysis by RP-HPLC of the WSN
fraction, were more effective than the Urea-PAGE electrophoresis analysis for discriminating
the samples according to its quality. And as a matter of fact, it was possible to observe
considerable differences in the primary and secondary proteolysis by means of these methods.
Soluble Nitrogen Fractions and TKN
The results from this analysis show differences in the primary and secondary protelolysis
related to the quality of the samples assessed. Thus from the fractionation scheme proposed
and employed by Ardö and Frederiksberg 1999, and Sousa, Ardö, and McSweeney 2001, it
can be seen in figures 65 to 69, that the nitrogen concentration increased during time and it
was higher for the good samples. The WSN includes all casein breakdown products, but
native caseins and high molecular weight peptides; the 12% trichloracetic TCA-SN contains
small peptides and FAA; and PTA-SN, which contains the smallest peptides (600 Da) and
FAA (T. M. I. E. Christensen, Bech, and Werner 1991).
In figures 65 and 66, it was confirmed that the nitrogen levels for the WSN fraction were the
highest in comparison to TCA-SN and PTA-SN fractions. Indeed, the WSN fraction had
values about 3 times larger than those of the TCA-SN fraction and 7 to 8 times higher in
comparison to those of the PTA-SN fraction. On th other hand, it is possible to see in figures
67, 68 and 69 that nitrogen content by the end of the observation for the WSN and TCA-SN
fractions was 1.5 to 2 times higher than those at the beginning of the observation, while in the
case of the PTA-SN fraction, the nitrogen values at the end were 2 to 2.5 times bigger than
those starting with.
In relation to the nitrogen levels of the WSN fraction in figure 67, it can be seen that the
values increased during ripening and were clearly different between samples from the very
beginning of the maturation process. Also, it is possible to appreciate that the results for the
“good” samples were higher than those for the “weak samples”. Therefore, based on the fact
that this fraction represents the primary proteolysis (Bansal, Piraino, and McSweeney 2009),
where the main enzymatic activity is proportionate by the rennet and the indigenous milk
proteinases (Allan J. Cliffe, Marks, and Mulholland 1993; A.J. Cliffe, Revell, and Law 1989),
it is possible to relate the difference in patterns of proteolysis to a marked disparity in
179
manufacturing operations that are capable to alterate the performance of these enzymes such
as: 1) standardization of milk (adjustments of milk composition, and pasteurization of cheese
milk), 2) pH achieved during acidification and at whey drainage, which determines the
retention of coagulant activity, 3) and moisture content of the curd; resulting in defects such as
sour and/or bitter flavor and soft and pasty body.
Concerning the nitrogen levels of the TCA-SN fraction, the figure 68 shows the same behavior
of the WSN fraction, where nitrogen levels increased during time and the “good” samples had
higher values. Also it can be seen that the evolution of this fraction results in final values that
are 1.5 bigger than the ones at the beginng of the observation. This fraction is rich in small
peptides of low and medium hydrophobicity (Kuchroo & Fox 1982), mostly resulting from
LAB, NSLAB and rennet enzymatic activity (Tanoj K. Singh, Fox, and Healy 1995; T.K.
Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox
1998), which is the reason why it is possible to attribute the difference found to inconsistency
in processing variables such as temperature and relative humidity in ageing rooms, heat
treatment of cheese milk, amount of salt added during salting and pH at salting, which affect
the intracellular and extracellular proteolityc activity proportionated by LAB and NSLAB.
As it is well known the PTA-SN fraction is an index of secondary proteolysis because it is
mainly constituted by very small peptides (<15 kDa) and amino acids of approximately 600
Da (Aston and Dulley 1982). The results in figure 69, showed a similar trend to the other
fractions; the values increased over time and were higher for the “good” samples. Even so,
the proportional increase in nitrogen for this fraction was more steadfast, and the final values
were about 2 to 3 times higher than those at the begin of the observation. Alike the WSN and
TCA-SN fractions, the difference between samples was evident since the beginning of the
observation period and as in the case of the TCA-SN fraction it could be explain through the
difference in processing variable that affect LAB and NSLAB
In general, it is possible to state that the fractionation scheme based on extraction of peptides
with water can establish a fair comparison between samples of different quality in order to
track and evaluate the extend of proteolysis.
180
Figure 65 TKN fractions Good cheese
Figure 66 TKN fractions Good cheese
181
Figure 67 WSN Good Vs Weak cheese
Figure 68 WSN Good Vs Weak cheese
182
Figure 69 PTA-SN Good Vs Weak cheese
Peptide analysis by RP-HPLC
The peptide analysis by RP-HPLC is another index of secondary proteolysis. In addition, it
can be used in authenticity studies and optimization process (Upadhyay et al. 2004).
In this work to evaluate the difference in the peptide profile for the different samples a
principal component analysis (PCA) was used, which is a multivariate analysis tool use in
descriptive statistics, to estimate the linear relationship between variables when their number
is very large (Chatfield and Collins 1981). As matter of fact, the data from chromatograms
was processed based on Piraino, Parente, and McSweeney 2004 work, where the complexity
of profiles containing more than 70 peaks is initially reduced using time intervals, whose area
is expressed as a percentage of total area of the chromatogram. Then the variability due the
characteristics of the samples in terms of treatments, biological factors (ripening, cheese
making process, milk quality, etc) and technical factors (sampling, extraction steps, and
measurement of peak and intervals area), is measured and analyzed through the contributing
eigenvalues in the correlation matrix for the principal components of the resulting model.
The peptide profiles revealed visible differences that were supported by the PCA results.
Indeed, the PCA analysis leads to a model with three principal components (PCs) that explain
183
the 80.7% of the variability of the data. However, after comparing in pairs the score plots for
any combination of the principal components, only the score plot for the PC1 Vs PC2 revealed
a correlation. This is in agreement with the published results of (Benfeldt and Sørensen 2001).
The first and second principal components explained the 72% of the variance, and the score
plots in figures 71 and 72 show how the model can differentiate the samples according to their
age in the PC1 and according their quality in PC2. Therefore, the figures display higher scores
for those samples with longer ripening that correspond to the good samples, which indicate
that proteolysis is faster for cheeses corresponding to the “good” samples.
The loading plot in figure 73, shows the projection of the eluting intervals on the PC1 and
PC2, and it allows to establish a correlations between the type of sample, its age and the
amount of peptides eluting within certain retention times. Thus, it can be seen that the
segments from 12-20 and 20-25 minutes have high scores for the PC1 and positive values
close to 0 for PC2. This trend suggests that the amount of peptides eluting in this zones rise
over time and higher amounts could be associate to good quality cheese. This interpretation
could be explain by the fact that the segments from 12-20 and 20-25 are mainly composed by
hydroplhilic peptides and free amino acids (FAA) resulting from the action of Chymosin on
αs1-CN and k-caseins, and the cell envelope proteinase (CEP) on the peptide αs1-CN(f1-23),
which accumulate during ripening (Tanoj K. Singh, Fox, and Healy 1997; Manuela
Fernández, Singh, and Fox 1998). Thus, based on the works of Singh and Fox 1998, which
describes the WSN fraction as the one that contains many compounds associated to the
characteristic savory flavor in cheese. It could be possible to suggest that higher amounts of
these hydrophilic peptides and free aminoacids might be related to “good” quality cheese.
Which can associated to processing variables such as …………. that direct and indirectly
affect proteolytic systems such as Chymosin and CEP.
Contrary, the segments 25-30 and 30-35 got negative loadings for PC1 and PC2, which
means that the relative amount of peptides eluting in this intervals decrease over time and
therefore higher amounts might be related to “weak” cheese. This is believed to do with the
breakdown of the αs1-CN, αs2-CN, αs1-CN (f24-199) and β-CN peptides, corresponding to
184
enzymatic activity proportionate by the rennet, indigenous milk enzymes and LAB enzymes
(T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh,
and Fox 1998); making sense since the breakdown of the main caseins is progressive and a
higher concentration of them could be related to the original character of curd or to a mild and
definitevily “non”-sharp Cheddar cheese. Nonetheless, in spite of the clear discrimination of
samples according to quality, it is not easy to explain the reason why, which is consequence of
the proteolytic systems involved in these segments that basically are all of them. This
indicates the obvious dissimilarity between the production process of these two types of
samples.
Regarding the intervals 35-40 and 40-45, they are mainly constituted by hydrophobic peptides
such as: 1) the fragments β-CN(f29–209), β-CN(f106–209) and β-CN(f108–209) (γ1, γ 2, and
γ 3, respectively), whose concentrations increase during ripening (Farkye and Fox 1990) and
are the result from the hydrolysis of β-Casein by Plasmin at Lys28-Lys29, Lys105-Gln106 and
Lys107-Glu108 bonds; 2) the peptides αs1-CN(f93–?), αs1-CN(f24–30), αs1-CN(f26–32),
αs1-CN(f26–34) resulting from the hydrolysis of the peptide αs1-CN(f24–199) by Chymosin,
CEP and aminopeptidase (T.K. Singh et al. 1994; Tanoj K. Singh, Fox, and Healy 1995; Tanoj
K. Singh, Fox, and Healy 1997; Manuela Fernández, Singh, and Fox 1998); 3) peptides αs2CN(f204–207), which is a C-terminal residue and product of lactococcal CEP (T.K. Singh et
al. 1994). Therefore, the positive scores for PC1 and the negative ones for PC2 means that the
amount of peptides eluting in these intervals increased during time and higher amounts are
related to “good” cheese. As in the case of the last two intervals, the development of these
peptides involved all proteolytic systems, thus it is difficult to point out a specific part of the
manufacturing process that causes the difference detected.
The intervals between 45-65 minutes are mainly constituted by hydrophobic compounds;
however, they displayed a different trend to that of the intervals between 35 to 45 min. The
scores in figure 73 showed negative values for the intervals 45-50, 50-55 and 55-60 in the
PC1 while the interval 60-65 got a small and positive score, which is close to zero. This means
that those the amount of peptides eluting in the first two intervals decreased during time
whereas the constituents of last interval increased. On the other hand the interval 45-50 and
50-55 got scores fairly close to zero for the PC2, which suggests that there is not a big
185
difference between samples for this part of the chromatograms. Contrary the interval 55-60
got a positive value in PC2, which indicates that the smaller is the amount of material eluting
in this interval the better the quality of the sample. The interval 60-65 got a negative score and
this suggests that the bigger is the amount of the material eluting in this zone the lower is the
quality of the samples. A possible explanation for the the way the samples were discriminated
by the model might has to do with the bitter character of some of the hydrophobic peptides
and free amino acids of this zone, which essentially are those that contain aromatic amino acid
(Allan J. Cliffe, Marks, and Mulholland 1993). Therefore the last two observations respect the
PC2 makes sense since bitterness is one of the defects to avoid in Cheddar cheese.
Figure 70 Scree plot of Good Vs Weak cheese
186
Figure 71 Score plot of Good Vs Weak cheese (age)
Figure 72 Score plot of Good Vs Weak cheese (quality)
187
Figure 74 Peptide profile Weak Cheese (A and B) Vs Good Cheese (C and D)
A
C
B
D
188
Electrophoresis
The results of the urea-PAGE of the water insoluble nitrogen fraction of experimental Cheddar
cheese in figures 75 and 76, clearly exhibit the progressive change of αS1-CN into the peptides
αS1-CN (f24-199), αS1-CN (f121-199), αS1-CN (f99-199), and the β-CN into peptides β-CN
(f29-202), β-CN (f108-209) and β-CN (f106-209) during ripening. In addition bands of γ-CNs
became more noticeable after 4 months. Also it can be seen that apparently the development
of peptides from αS1-CN is faster than those from β-CN, which can be related to primary
proteolysis and the actual amount of β-CN that is hydrolyzed (only the 50%). However, the
representative urea-PAGE gel do not show any appreciable difference between samples of
different quality.
Figure 75 Urea PAGE for ripening of Good cheese
β -CN(f106-209)
β -CN(f29-209)
β -CN(f108-209)
β -CN
αs1 -CN
αs1 -CN(f24-199)
αs1-CN(f121-199)
2M
6M
9M
12M
15MG
15MW
189
Figure 76 77 Urea PAGE for Good Vs Weak cheese (12 months)
Good
Weak
CONCLUSION
The present study demonstrates that the use of FFA profile, VSC’s profile, measurement of
the levels of the Total Kjeldahl nitrogen for the WSN, TCA-SN and PTA-SN fractions, and
the PCA of the RP-HPLC peptide profile of the WSN fraction are effective tools and
ripening indices to differentiate Cheddar cheese samples regarding their age and quality. The
urea-PAGE was
effective to differentiate samples by their age; nonetheless it was not
sensitive enough to detect differences related to quality. On the other hand, , the results of
levels of nitrogen for all the 3 fractions analyzed demonstrated that proteolysis is faster for
good cheeses This was supported by the PCA model obtained which suggests many possible
causes. Lipolysis was slower for weak cheese, which showed lower levels of individual FFA.
The amounts of DMS, H2S and MeSH showed noticeable differences between samples and it
can be seen that good cheese undergoes a faster catabolism of sulfur containing. Once again,
the results for DMDS and DMTS suggest that they are artifacts from extraction and separation
procedures rather than metabolites from the ripening of Cheddar cheese.
190
REFERENCES
Alting, A. C., W. Engels, S. van Schalkwijk, and F. A. Exterkate. 1995. “Purification and
Characterization of Cystathionine (beta)-Lyase from Lactococcus Lactis Subsp.
Cremoris B78 and Its Possible Role in Flavor Development in Cheese.” Applied and
Environmental Microbiology V61 (11) (November 1): pp. 4037–4042.
Amarita, F., T. Requena, G. Taborda, L. Amigo, and C. Pelaez. 2001. “Lactobacillus Casei
and Lactobacillus Plantarum Initiate Catabolism of Methionine by Transamination.”
Journal of Applied Microbiology V90 (6): pp. 971–978.
Aston, J. W., and L. K. Creamer. 1986. “Contribution of the components of the water-soluble
fraction to the flavour of Cheddar cheese.” New Zealand Journal of Dairy Science and
Technology V21 (3): pp. 229–248.
Aston, J. W., and John R. Dulley. 1982. “Cheddar cheese flavor review.” Australian Journal
of Dairy Technology V37 (2): pp. 59–64.
Baer, A., and I. Ryba. 1992. “Serological Identification of Propionibacteria in Milk and
Cheese Samples.” International Dairy Journal V2 (5): pp. 299–310.
Balcão, Victor M., and F.Xavier Malcata. 1998. “Lipase Catalyzed Modification of Milkfat.”
Biotechnology Advances V16 (2) (March): pp. 309–341.
Bansal, N, Paolo Piraino, and P. L. H. McSweeney. 2009. “Determination of Proteolysis in
Cheese.” In Handbook of Dairy Foods Analysis, pp. 405–417. CRC Press.
Benfeldt, Connie, and John Sørensen. 2001. “Heat Treatment of Cheese Milk: Effect on
Proteolysis During Cheese Ripening.” International Dairy Journal V11 (4–7): pp.
567–574.
Bentley, Ronald, and Thomas G. Chasteen. 2004. “Environmental VOSCs––formation and
Degradation of Dimethyl Sulfide, Methanethiol and Related Materials.” Chemosphere
V55 (3): pp. 291–317.
Bills, D.D., and E.A. Day. 1964. “Determination of the Major Free Fatty Acids of Cheddar
Cheese.” Journal of Dairy Science V47 (7): pp. 733–738.
Burbank, Helen M., and Michael C. Qian. 2005. “Volatile Sulfur Compounds in Cheddar
Cheese Determined by Headspace Solid-phase Microextraction and Gas
Chromatograph-pulsed Flame Photometric Detection.” Journal of Chromatography A
V1066 (1–2): pp. 149–157.
Burbank, Helen, and Michael C. Qian. 2008. “Development of Volatile Sulfur Compounds in
Heat-shocked and Pasteurized Milk Cheese.” International Dairy Journal V18 (8)
(August): pp. 811–818.
Chavarri, Felisa, Mailo Virto, Celia Martin, Ana I. Nájera, Arantza Santisteban, Luis J. R.
Barrón, and Mertxe De Renobales. 1997. “Determination of Free Fatty Acids in
Cheese: Comparison of Two Analytical Methods.” Journal of Dairy Research V64
(03): pp. 445–452.
Christensen, Tove M. I. E., Kristian R. Kristiansen, and Jesper S. Madsen. 1989. “Proteolysis
in Cheese Investigated by High Performance Liquid Chromatography.” Journal of
Dairy Research V56 (05): pp. 823–828.
Cliffe, A.J., D. Revell, and B.A. Law. 1989. “A Method for the Reverse Phase High
Performance Liquid Chromatography of Peptides from Cheddar Cheese.” Food
Chemistry 34 (2): 147–160.
191
Cliffe, Allan J., Jonathan D. Marks, and Francis Mulholland. 1993. “Isolation and
Characterization of Non-volatile Flavours from Cheese: Peptide Profile of Flavour
Fractions from Cheddar Cheese, Determined by Reverse-phase High-performance
Liquid Chromatography.” International Dairy Journal V3 (4–6): pp 379–387.
Collins, Yvonne F., Paul L.H. McSweeney, and Martin G. Wilkinson. 2003. “Lipolysis and
Free Fatty Acid Catabolism in Cheese: a Review of Current Knowledge.”
International Dairy Journal V13 (11): pp. 841–866.
Collins, Y.F., P.L.H. McSweeney, and M.G. Wilkinson. 2004. “Lipolysis and Catabolism of
Fatty Acids in Cheese.” In Cheese Chemistry, Physics and Microbiology, Volume
1:373–389. Academic Press.
Curioni, P.M.G., and J.O. Bosset. 2002. “Key Odorants in Various Cheese Types as
Determined by Gas Chromatography-olfactometry.” International Dairy Journal v12
(12): pp. 959–984.
Datta, N., A. J. Elliott, M. L. Perkins, and H. C. Deeth. 2002. “Ultra-high-temperature (UHT)
Treatment of Milk: Comparison of Direct and Indirect Modes of Heating” V57 (3):
pp. 211–227.
Deeth, H. C., and V. Touch. 2000. “Methods for detecting lipase activity in milk and milk
products.” Australian Journal of Dairy Technology V55 (3): pp. 153–168.
Dias, Benjamin, and Bart Weimer. 1998. “Conversion of Methionine to Thiols by Lactococci,
Lactobacilli, and Brevibacteria.” Applied and Environmental Microbiology V64 (9):
pp. 3320–3326.
Farkye, Nana Y., and Patrick F. Fox. 1990. “Observations on Plasmin Activity in Cheese.”
Journal of Dairy Research V57 (03): pp. 413–418.
Fennema, Owen R., and Damodaran. 1996. “Amino Acids, Peptides, and Proteins.” In Food
Chemistry, pp. 322–425. CRC Press.
Fernández, Manuela, Tanoj K. Singh, and Patrick F. Fox. 1998. “Isolation and Identification
of Peptides from the Diafiltration Permeate of the Water-Soluble Fraction of Cheddar
Cheese.” Journal of Agricultural and Food Chemistry V46 (11): pp. 4512–4517.
Fernández-García, Estrella. 1996. “Use of Headspace Sampling in the Quantitative Analysis
of Artisanal Spanish Cheese Aroma.” Journal of Agricultural and Food Chemistry
V44 (7): pp. 1833–1839.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Fox, P.F. 1989. “Proteolysis During Cheese Manufacture and Ripening.” Journal of Dairy
Science V72 (6): pp. 1379–1400.
Fox, P.F., and L. Stepaniak. 1993. “Enzymes in Cheese Technology.” International Dairy
Journal V3 (4–6): pp. 509–530..
Fox, P. F., and P. L. H. McSweeney. 1996. “Proteolysis in Cheese During Ripening.” Food
Reviews International V12 (4): pp. 457–509.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Gao, Song, Ed S. Mooberry, and James L. Steele. 1998. “Use of 13C Nuclear Magnetic
Resonance and Gas Chromatography To Examine Methionine Catabolism by
Lactococci.” Applied and Environmental Microbiology V64 (12): pp. 4670–4675.
Green, Margaret L., and Donald J. Manning. 1982. “Development of Texture and Flavour in
Cheese and Other Fermented Products.” Journal of Dairy Research V49 (04): pp.
737–748.
192
Gripon, Jean-Claude, V Monnet, G Lambert, and M. J. Desmazeaud. 1991. “Microbial
enzymes in cheese ripening.” In Food enzymology, vol 1.
K. Arfi, H.Spinnler, R. Tache, and P. bonnarme. 2002. “Production of Volatile Compounds by
Cheese-ripening Yeasts: Requirement for a Methanethiol Donor for S-methyl
Thioacetate Synthesis by Kluyveromyces Lactis.” Applied Microbiology and
Biotechnology V58 (4): pp. 503–510.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen. 1991. “Influence of
Pasteurization of Milk on Protein Breakdown in Cheddar Cheese During Aging.”
Journal of Dairy Science V74 (3): pp. 727–740.
Lee, Won-Jae, Dattatreya S. Banavara, Joanne E. Hughes, Jason K. Christiansen, James L.
Steele, Jeffery R. Broadbent, and Scott A. Rankin. 2007. “Role of Cystathionine βLyase in Catabolism of Amino Acids to Sulfur Volatiles by Genetic Variants of
Lactobacillus Helveticus CNRZ 32.” Applied and Environmental Microbiology V73
(9): pp. 3034–3039.
Manning, D. J., Helen R. Chapman, and Zena D. Hosking. 1976. “The Production of Sulphur
Compounds in Cheddar Cheese and Their Significance in Flavour Development.”
Journal of Dairy Research V43 (2): pp. 313–320.
Manning, Donald J., and Carolyn Moore. 1979. “Headspace Analysis of Hard Cheeses.”
Journal of Dairy Research V46 (3): pp. 539–545.
McSweeney, P.L.H., P.F. Fox, J.A. Lucey, K.N. Jordan, and T.M. Cogan. 1993. “Contribution
of the Indigenous Microflora to the Maturation of Cheddar Cheese.” International
Dairy Journal V3 (7): pp. 613–634.
McSweeney, Paul L. H., Sylvie Pochet, Patrick F. Fox, and Aine Healy. 1994. “Partial
Identification of Peptides from the Water-Insoluble Fraction of Cheddar Cheese.”
Journal of Dairy Research V61 (4) (04): pp. 587–590.
McSweeney, Paul L.H., and Maria José Sousa. 2000. “Biochemical Pathways for the
Production of Flavour Compounds in Cheeses During Ripening: A Review.” Le Lait
V80 (3): pp. 293–324.
Metwalli, Ali A.M, Harmen H.J de Jongh, and Martinus A.J.S van Boekel. 1998. “Heat
Inactivation of Bovine Plasmin.” International Dairy Journal V8 (1): pp. 47–56.
Milo, C., and G. A. Reineccius. 1997. “Identification and Quantification of Potent Odorants in
Regular-Fat and Low-Fat Mild Cheddar Cheese.” Journal of Agricultural and Food
Chemistry V45 (9) (September 1): pp. 3590–3594.
Molimard, P., and H.E. Spinnler. 1996. “Review: Compounds Involved in the Flavor of
Surface Mold-Ripened Cheeses: Origins and Properties.” Journal of Dairy Science
V79 (2) (February): pp. 169–184.
O’Keeffe, Arthur M., Patrick F. Fox, and Charles Daly. 1978. “Proteolysis in Cheddar Cheese:
Role of Coagulant and Starter Bacteria.” Journal of Dairy Research V45 (3): pp. 465–
477..
O’Keeffe, R. B., P. F. Fox, and C. Daly. 1976. “Contribution of Rennet and Starter Proteases
to Proteolysis in Cheddar Cheese.” Journal of Dairy Research 43 (1): 97–107.
Piraino, Paolo, Eugenio Parente, and Paul L. H. McSweeney. 2004. “Processing of
Chromatographic Data for Chemometric Analysis of Peptide Profiles from Cheese
Extracts:  A Novel Approach.” Journal of Agricultural and Food Chemistry 52 (23):
6904–6911.
Preininger, Martin, and Werner Grosch. 1994. “Evaluation of Key Odorants of the Neutral
Volatiles of Emmentaler Cheese by the Calculation of Odour Activity Values.” LWT Food Science and Technology V27 (3): pp. 237–244.
193
Qian, M., and G. Reineccius. 2002. “Identification of Aroma Compounds in ParmigianoReggiano Cheese by Gas Chromatography/Olfactometry.” Journal of Dairy Science
V85 (6) pp. 1362–1369.
Seefeldt, K.E., and B.C. Weimer. 2000. “Diversity of Sulfur Compound Production in Lactic
Acid Bacteria.” Journal of Dairy Science V83 (12): pp. 2740–2746.
Singh, T. K, M. A Drake, and K. R Cadwallader. 2003. “Flavor of Cheddar Cheese: A
Chemical and Sensory Perspective.” Comprehensive Reviews in Food Science and
Food Safety V2 (4): pp. 166–189.
Singh, T.K., P.F. Fox, P. Højrup, and A. Healy. 1994. “A Scheme for the Fractionation of
Cheese Nitrogen and Identification of Principal Peptides.” International Dairy
Journal V4 (2): pp. 111–122.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1995. “Water-Soluble Peptides in Cheddar
Cheese: Isolation and Identification of Peptides in the Diafiltration Retentate of the
Water-Soluble Fraction.” Journal of Dairy Research V62 (4): pp. 629–640.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1997. “Isolation and Identification of Further
Peptides in the Diafiltration Retentate of the Water-soluble Fraction of Cheddar
Cheese.” Journal of Dairy Research V64 (3): pp. 433–443.
Smacchi, Emanuele, and Marco Gobbetti. 1998. “Purification and Characterization of
Cystathionine Γ-lyase from Lactobacillus Fermentum DT41.” FEMS Microbiology
Letters V166 (2): pp. 197–202.
Smith, A. M., and S. Nakai. 1990. “Classification of cheese varieties by multivariate analysis
of HPLC profiles.” Canadian Institute of Food Science and Technology Journal V23
(1): pp. 53–58.
Sousa, M.J, Y Ardö, and P.L.H McSweeney. 2001. “Advances in the Study of Proteolysis
During Cheese Ripening.” International Dairy Journal V11 (4-7): pp. 327–345.
Subramanian, A., W.J. Harper, and L.E. Rodriguez-Saona. 2009. “Cheddar Cheese
Classification Based on Flavor Quality Using a Novel Extraction Method and Fourier
Transform Infrared Spectroscopy.” Journal of Dairy Science V92 (1): pp. 87–94.
Tanaka, Hidehiko, Nobuyoshi Esaki, and Kenji Soda. 1985. “A Versatile Bacterial Enzyme:
L-methionine Γ-lyase.” Enzyme and Microbial Technology V7 (11) (November): pp.
530–537.
Urbach, G. 1995. “Contribution of Lactic Acid Bacteria to Flavour Compound Formation in
Dairy Products.” International Dairy Journal V5 (8): pp. 877–903.
USDA. 2011. Dairy Products Annual Sumary.
Visser, Servaas. 1993. “Proteolytic Enzymes and Their Relation to Cheese Ripening and
Flavor: An Overview.” Journal of Dairy Science V76 (1): pp. 329–350.
Voigt, Daniela D., François Chevalier, John A. Donaghy, Margaret F. Patterson, Michael C.
Qian, and Alan L. Kelly. 2012. “Effect of High-pressure Treatment of Milk for
Cheese Manufacture on Proteolysis, Lipolysis, Texture and Functionality of Cheddar
Cheese During Ripening.” Innovative Food Science & Emerging Technologies V13
(0): pp. 23–30.
Weimer, Bart, Kimberly Seefeldt, and Benjamin Dias. 1999. “Sulfur Metabolism in Bacteria
Associated with Cheese.” Antonie Van Leeuwenhoek 76 (1): 247–261.
de Wit, Rudy, and Hans Nieuwenhuijse. 2008. “Kinetic Modelling of the Formation of
Sulphur-containing Flavour Components During Heat-treatment of Milk.”
International Dairy Journal V18 (5): pp. 539–547.
Yvon, Mireille, and Liesbeth Rijnen. 2001. “Cheese Flavour Formation by Amino Acid
Catabolism.” International Dairy Journal V11 (4–7): pp. 185–201.
194
195
CONCLUSION
The present study demonstrates that the use of FFA profile, VSC’s profile, measurement of
the level of the Total Kjeldahl nitrogen for the WSN, TCA-SN and PTA-SN fractions, and the
PCA analysis of the RP-HPLC peptide profile of the WSN fraction are effective tools and
ripening indices to differentiate Cheddar cheese samples regarding variables such as quality,
age, heat treatment of milk, origin, and addition of adjunct culture. In the case of Urea-PAGE,
it was demonstrated that its use as index of primary proteolysis can be effective to differentiate
samples by their age; nonetheless it is not sensitive enough to detect differences related to the
changes caused by heat treatment of cheese milk, origin of samples and addition of adjunct
culture.
It was demonstrated that proteolysis is faster for good cheese and cheeses made with adjunct
culture, made in the TCCA plant and made from heat-shocked milk. In general the levels of
nitrogen for the 3 fractions analyzed were the highest for these treatments. The results were
supported by the PCA models obtained, which suggest differences caused by: 1) the role of
the adjunct culture as supplement to the starter culture during ripening, 2) distinct
manufacturing practices between prodcutction plants, and 3) physical changes to milk
structure and population of NSLAB as consequence of heat treatments of cheese milk.
Llipolysis is slower for weak samples, cheese produce without adjunct culture, cheese made
with pasteurized milk and for samples made in the CRP plant, which showed lower levels of
individual FFA, speacially for short chain fatty acids. Finally the amounts of DMS, H2S and
MeSH, revealed differences between the treatments, and it is possible to appreciate the
tendency to accelerate the catabolism of sulfur containing amino acids, such as methionine
and cysteine, for the samples with adjunct culture, made from heat-shocked milk and from the
TCCA plant. An important outcome from this research work was to point out the possibility
that DMDS and DMTS are artifacts from extraction and separation procedures rather than
metabolites from the ripening of Cheddar cheese.
Other tests such as analysis of amino acids or measurement of the levels of other potent
volatile compounds through GC-MS should be performed in order to complement the
proposed chemical analysis and to obtain more information to keep explaining the causes of
196
the difference in rates and patterns of lipolysis and proteolysis. In addition, it would be usuful
to totally assure that the same manufacturing procedures and raw materials are used during the
manufacture of samples, which might give a better idea about the influence of the investigated
parameters in the final results by eliminating the noise of other manufacturing variables.
Other areas that deserve more research and are not well understood yet, are related to the
specific role of important strains of NSLAB in lipolysis and proteolysis, and the generation of
DMS and H2S.
197
BIBLIOGRAFY
Alichanidis, Efstathios, Julia H. M. Wrathall, and Anthony T. Andrews. 1986. “Heat Stability
of Plasmin (milk Proteinase) and Plasminogen.” Journal of Dairy Research V53 (2):
pp. 259–269.
Alting, A. C., W. Engels, S. van Schalkwijk, and F. A. Exterkate. 1995. “Purification and
Characterization of Cystathionine (beta)-Lyase from Lactococcus Lactis Subsp.
Cremoris B78 and Its Possible Role in Flavor Development in Cheese.” Applied and
Environmental Microbiology V61 (11): pp. 4037–4042.
Amarita, F., T. Requena, G. Taborda, L. Amigo, and C. Pelaez. 2001. “Lactobacillus Casei
and Lactobacillus Plantarum Initiate Catabolism of Methionine by Transamination.”
Journal of Applied Microbiology V90 (6): pp. 971–978.
Andersen, Lene T., Ylva Ardö, and Wender L.P. Bredie. 2010. “Study of Taste-active
Compounds in the Water-soluble Extract of Mature Cheddar Cheese.” International
Dairy Journal V20 (8): pp. 528–536.
Anderson, D.L., V.V. Mistry, R.L. Brandsma, and K.A. Baldwin. 1993. “Reduced Fat
Cheddar Cheese from Condensed Milk. 1. Manufacture, Composition, and Ripening.”
Journal of Dairy Science V76 (10): pp. 2832–2844.
Ardö, Y., and Frederiksberg. 1999. “Evaluating Proteolysis by Analysing the N Content of
Cheese Fractions.” Bulletin - International Dairy Federation V337: pp. 4–9.
Arora, Gulshan, Francois Cormier, and Byong Lee. 1995. “Analysis of Odor-Active Volatiles
in Cheddar Cheese Headspace by Multidimensional GC/MS/Sniffing.” J. Agric. Food
Chem. V43 (3): pp. 748–752.
Aston, J. W., and L. K. Creamer. 1986. “Contribution of the components of the water-soluble
fraction to the flavour of Cheddar cheese.” New Zealand Journal of Dairy Science and
Technology V21 (3): pp. 229–248.
Aston, J. W., and John R. Dulley. 1982. “Cheddar cheese flavor review.” Australian Journal
of Dairy Technology V37 (2): pp. 59–64.
Avsar, Y.K., Y. Karagul-Yuceer, M.A. Drake, T.K. Singh, Y. Yoon, and K.R. Cadwallader.
2004. “Characterization of Nutty Flavor in Cheddar Cheese.” Journal of Dairy
Science V87 (7): pp. 1999–2010.
Ayad, Eman H.E, Annette Verheul, Catrienus de Jong, Jan T.M Wouters, and Gerrit Smit.
1999. “Flavour Forming Abilities and Amino Acid Requirements of Lactococcus
Lactis Strains Isolated from Artisanal and Non-dairy Origin.” International Dairy
Journal V9 (10): pp. 725–735.
Baer, A., and I. Ryba. 1992. “Serological Identification of Propionibacteria in Milk and
Cheese Samples.” International Dairy Journal V2 (5): pp. 299–310.
Balcão, Victor M., and F.Xavier Malcata. 1998. “Lipase Catalyzed Modification of Milkfat.”
Biotechnology Advances V16 (2) (March): pp. 309–341.
Bansal, N, Paolo Piraino, and P. L. H. McSweeney. 2009. “Determination of Proteolysis in
Cheese.” In Handbook of Dairy Foods Analysis, pp. 405–417. CRC Press.
Bassett, Harold J., and W.L. Slatter. 1953. “The Heat Resistance of Some Lactobacilli Found
in Swiss Cheese.” Journal of Dairy Science V36 (4): pp. 409–413.
Bastian, Eric D., and Rodney J. Brown. 1996. “Plasmin in Milk and Dairy Products: An
Update.” International Dairy Journal V6 (5): pp. 435–457.
198
Bastian, E.D., C.G. Lo, and K.M.M. David. 1997. “Plasminogen Activation in Cheese Milk:
Influence on Swiss Cheese Ripening.” Journal of Dairy Science 80 (2): pp. 245–251.
Benfeldt, Connie, John Sørensen, Katrine H. Ellegård, and Torben E. Petersen. 1997. “Heat
Treatment of Cheese Milk: Effect on Plasmin Activity and Proteolysis During Cheese
Ripening.” International Dairy Journal 7 (11): 723–731.
Benfeldt, Connie, and John Sørensen. 2001. “Heat Treatment of Cheese Milk: Effect on
Proteolysis During Cheese Ripening.” International Dairy Journal V11 (4–7): pp.
567–574.
Benkler, K. F., and G. A. Reineccius. 1980. “Flavor isolation from fatty foods via solvent
extraction and membrane dialysis.” Journal of Food Science V45 (4): pp. 1084–1085.
Bentley, Ronald, and Thomas G. Chasteen. 2004. “Environmental VOSCs––formation and
Degradation of Dimethyl Sulfide, Methanethiol and Related Materials.” Chemosphere
V5 (3): pp. 291–317.
Beresford, Tom P, Nora A Fitzsimons, Noelle L Brennan, and Tim M Cogan. 2001. “Recent
Advances in Cheese Microbiology.” International Dairy Journal V11 (4–7): pp. 259–
274.
Beuvier, E., and S. Buchin. 2004. “Raw Milk Cheeses.” In Cheese Chemistry, Physics and
Microbiology, Volume 1:319–345. Academic Press.
Bills, D.D., and E.A. Day. 1964. “Determination of the Major Free Fatty Acids of Cheddar
Cheese.” Journal of Dairy Science V47 (7): pp. 733–738.
Burbank, Helen M., and Michael C. Qian. 2005. “Volatile Sulfur Compounds in Cheddar
Cheese Determined by Headspace Solid-phase Microextraction and Gas
Chromatograph-pulsed Flame Photometric Detection.” Journal of Chromatography A
V1066 (1–2): pp. 149–157.
Burbank, Helen, and Michael C. Qian. 2008. “Development of Volatile Sulfur Compounds in
Heat-shocked and Pasteurized Milk Cheese.” International Dairy Journal V18 (8):
pp. 811–818.
Burton, H. 1986. “Microbiological aspects.” V200: pp. 9–14.
Careri, M, and A Mangia. 2003. “Analysis of Food Proteins and Peptides by Chromatography
and Mass Spectrometry.” Journal of Chromatography A V1000 (1–2): pp. 609–635.
Carunchia Whetstine, Mary E., Keith R. Cadwallader, and MaryAnne Drake. 2005.
“Characterization of Aroma Compounds Responsible for the Rosy/Floral Flavor in
Cheddar Cheese.” J. Agric. Food Chem. V53 (8): pp. 3126–3132.
Carunchia Whetstine, M E, P J Luck, M A Drake, E A Foegeding, P D Gerard, and D M
Barbano. 2007. “Characterization of Flavor and Texture Development Within Large
(291 Kg) Blocks of Cheddar Cheese.” Journal of Dairy Science V90 (7): pp. 3091–
3109.
Chaintreau, Alain. 2001. “Simultaneous Distillation–extraction: From Birth to Maturity—
review.” Flavour and Fragrance Journal V16 (2): pp. 136–148.
Chatfield, Chris, and A. Collins. 1981. “Finding New Underlying Variables.” In Introduction
to Multivariate Analysis, pp. 57–78. CRC Press.
Chavarri, Felisa, Mailo Virto, Celia Martin, Ana I. Nájera, Arantza Santisteban, Luis J. R.
Barrón, and Mertxe De Renobales. 1997. “Determination of Free Fatty Acids in
Cheese: Comparison of Two Analytical Methods.” Journal of Dairy Research V64
(3): pp. 445–452.
Chich, Jean-François, Karine Marchesseau, and Jean-Claude Gripon. 1997. “Intracellular
Esterase from Lactococcus Lactis Subsp. Lactis NCDO 763: Purification and
199
Characterization.” International Dairy Journal V7 (2-3) (2–3) (February): pp. 169–
174.
Chopin, Alain. 1993. “Organization and Regulation of Genes for Amino Acid Biosynthesis in
Lactic Acid Bacteria.” FEMS Microbiology Reviews V12 (1–3): pp. 21–37.
Christensen J.E., Johnson M.E., and Steele J.L. 1995. “Production of Cheddar Cheese Using a
Lactococcus Lactis Ssp. Cremoris SK11 Derivative with Enhanced Aminopeptidase
Activity.” International Dairy Journal V5 (4): pp. 367–379.
Christensen, Jeffrey E., Edward G. Dudley, Jeffrey A. Pederson, and James L. Steele. 1999.
“Peptidases and Amino Acid Catabolism in Lactic Acid Bacteria.” Antonie Van
Leeuwenhoek V76 (1): pp. 217–246.
Christensen, K.R., and G.A. Reineccius. 1992. “Gas Chromatographic Analysis of Volatile
Sulfur Compounds from Heated Milk Using Static Headspace Sampling.” Journal of
Dairy Science V75 (8): pp. 2098–2104.
Christensen, Søren, Tove Wiegers, John Hermansen, and Lars Sottrup-Jensen. 1995. “Plasmaderived Protease Inhibitors in Bovine Milk.” International Dairy Journal V5 (5): pp.
439–449.
Christensen, Tove M. I. E., Kristian R. Kristiansen, and Jesper S. Madsen. 1989. “Proteolysis
in Cheese Investigated by High Performance Liquid Chromatography.” Journal of
Dairy Research V56 (5): pp. 823–828.
Christensen, T. M. I. E., A.-M. Bech, and H. Werner. 1991. “Methods for crude fractionation
(extraction and precipitation) of nitrogen components in cheese.” Bulletin International Dairy Federation (261): pp. 4–9.
Cliffe, A.J., D. Revell, and B.A. Law. 1989. “A Method for the Reverse Phase High
Performance Liquid Chromatography of Peptides from Cheddar Cheese.” Food
Chemistry 34 (2): 147–160.
Cliffe, Allan J., Jonathan D. Marks, and Francis Mulholland. 1993. “Isolation and
Characterization of Non-volatile Flavours from Cheese: Peptide Profile of Flavour
Fractions from Cheddar Cheese, Determined by Reverse-phase High-performance
Liquid Chromatography.” International Dairy Journal V3 (4–6): pp 379–387.
Collin, J. C., D. le Bars, I. Ryba, and A. Bear. 1990. “Comparative measurements of the
plasminogen-plasmin system of milk.” In International Dairy Federation.
Collins, Yvonne F., Paul L.H. McSweeney, and Martin G. Wilkinson. 2003. “Lipolysis and
Free Fatty Acid Catabolism in Cheese: a Review of Current Knowledge.”
International Dairy Journal V13 (11): pp. 841–866.
Collins, Y.F., P.L.H. McSweeney, and M.G. Wilkinson. 2004. “Lipolysis and Catabolism of
Fatty Acids in Cheese.” In Cheese Chemistry, Physics and Microbiology, Volume
1:373–389. Academic Press.
Considine, T., A. Healy, A.L. Kelly, and P.L.H. McSweeney. 1999. “Proteolytic Specificity of
Elastase on Bovine Β-casein.” Food Chemistry V66 (4): pp. 463–470..
Considine, T., Á. Healy, A.L. Kelly, and P.L.H. McSweeney. 2000. “Proteolytic Specificity of
Elastase on Bovine Αs1-casein.” Food Chemistry V69 (1): pp. 19–26.
Considine, T., A. Healy, A.L. Kelly. 2004. “Hydrolysis of Bovine Caseins by Cathepsin B, a
Cysteine Proteinase Indigenous to Milk.” International Dairy Journal V14 (2): pp.
117–124.
Crow, V.L., T. Coolbear, P.K. Gopal, F.G. Martley, L.L. McKay, and H. Riepe. 1995. “The
Role of Autolysis of Lactic Acid Bacteria in the Ripening of Cheese.” International
Dairy Journal V5 (8): pp. 855–875.
200
Crow, Vaughan, Brian Curry, and Michelle Hayes. 2001. “The Ecology of Non-starter Lactic
Acid Bacteria (NSLAB) and Their Use as Adjuncts in New Zealand Cheddar.”
International Dairy Journal V11 (4-7): pp 275–283.
Cserháti, Tibor. 2002. “Mass Spectrometric Detection in Chromatography. Trends and
Perspectives.” Biomedical Chromatography V16 (5): pp. 303–310.
Curioni, P.M.G., and J.O. Bosset. 2002. “Key Odorants in Various Cheese Types as
Determined by Gas Chromatography-olfactometry.” International Dairy Journal v12:
pp. 959–984.
Datta, N., A. J. Elliott, M. L. Perkins, and H. C. Deeth. 2002. “Ultra-high-temperature (UHT)
Treatment of Milk: Comparison of Direct and Indirect Modes of Heating” V57 (3):
pp. 211–227.
Deeth, H. C, and V. Touch. 2000. “Methods for detecting lipase activity in milk and milk
products.” Australian Journal of Dairy Technology V5 (3): pp. 153–168.
Dias, Benjamin, and Bart Weimer. 1998. “Conversion of Methionine to Thiols by Lactococci,
Lactobacilli, and Brevibacteria.” Applied and Environmental Microbiology V64 (9):
pp. 3320–3326.
Dimos, Tony. 1992. “A comparative study by GC/MS of the flavour volatiles produced during
the maturation of full-fat and low-fat Cheddar cheese.”
Dirinck, P, and A De Winne. 1999. “Flavour Characterisation and Classification of Cheeses
by Gas Chromatographic–mass Spectrometric Profiling.” Journal of Chromatography
A V847 (1–2): pp. 203–208.
Dobric, Nada, Gäetan K.Y. Limsowtin, Alan J. Hillier, Nicholas P.B. Dudman, and Barrie E.
Davidson. 2000. “Identification and Characterization of a Cystathionine Β/γ-lyase
from Lactococcus Lactis Ssp. Cremoris MG13631.” FEMS Microbiology Letters
V182 (2): pp. 249–254.
Drake, M.A., T.D. Boylston, K.D. Spence, and B.G. Swanson. 1996. “Chemical and Sensory
Effects of a Lactobacillus Adjunct in Cheddar Cheese.” Food Research International
V29 (3–4): pp. 381–387.
El Soda, Morsi A. 1993. “The Role of Lactic Acid Bacteria in Accelerated Cheese Ripening.”
FEMS Microbiology Reviews V12 (1–3): pp. 239–251.
El Soda, M., S.A. Madkor, and P.S. Tong. 2000. “Adjunct Cultures: Recent Developments
and Potential Significance to the Cheese Industry.” Journal of Dairy Science V83 (4):
pp. 609–619.
El-Soda, M., S. A. Madkor, and P. S. Tong. 2000. “Evaluation of commercial adjuncts for use
in cheese ripening: 4. Comparison between attenuated and not attenuated lactobacilli.”
Milchwissenschaft V55 (5): pp. 260–263.
Engel, Wolfgang, Wolfgang Bahr, and P. Schieberle. 1999. “Solvent Assisted Flavour
Evaporation – a New and Versatile Technique for the Careful and Direct Isolation of
Aroma Compounds from Complex Food Matrices.” European Food Research and
Technology V209 (3): pp. 237–241.
Engels, W.J.M., R. Dekker, C. de Jong, R. Neeter, and S. Visser. 1997. “A Comparative Study
of Volatile Compounds in the Water-soluble Fraction of Various Types of Ripened
Cheese.” International Dairy Journal V7 (4): pp. 255–263.
Exterkate, Fred, Arno C. Alting, and Charles J. Slangen. 1995. “Conversion of αs1-Casein(24-199)-Fragment and β-Casein Under Cheese Conditions by Chymosin and Starter
Peptidases.” Systematic and Applied Microbiology V18 (1): pp. 7–12.
201
Fang, Yu, and Michael C. Qian. 2005. “Sensitive Quantification of Sulfur Compounds in
Wine by Headspace Solid-phase Microextraction Technique.” Journal of
Chromatography A V1080 (2): pp. 177–185.
Farkye, Nana Y., and Patrick F. Fox. 1990. “Observations on Plasmin Activity in Cheese.”
Journal of Dairy Research V57 (3): pp. 413–418.
FDA. 2011. “Code of Federal Regulations Title 21.”
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=133.111.
Fennema, Owen R., and Damodaran. 1996. “Amino Acids, Peptides, and Proteins.” In Food
Chemistry, pp. 322–425. CRC Press.
Fernández, Manuela, Tanoj K. Singh, and Patrick F. Fox. 1998. “Isolation and Identification
of Peptides from the Diafiltration Permeate of the Water-Soluble Fraction of Cheddar
Cheese.” Journal of Agricultural and Food Chemistry V46 (11): pp. 4512–4517.
Fernández, María, Wim van Doesburg, Ger A. M. Rutten, Joey D. Marugg, Arno C. Alting,
Richard van Kranenburg, and Oscar P. Kuipers. 2000. “Molecular and Functional
Analyses of themetC Gene of Lactococcus Lactis, Encoding Cystathionine β-Lyase.”
Applied and Environmental Microbiology V66 (1): pp. 42–48..
Fernández-García, Estrella. 1996. “Use of Headspace Sampling in the Quantitative Analysis
of Artisanal Spanish Cheese Aroma.” Journal of Agricultural and Food Chemistry
V44 (7): pp. 1833–1839.
Ferreira, A.D., and B.C. Viljoen. 2003. “Yeasts as Adjunct Starters in Matured Cheddar
Cheese.” International Journal of Food Microbiology V86 (1-2): pp. 131–140.
Forde, Amanda, and Gerald F Fitzgerald. 2000. “Biotechnological Approaches to the
Understanding and Improvement of Mature Cheese Flavour.” Current Opinion in
Biotechnology V11 (5): pp. 484–489.
Fox, P.F. 1989. “Proteolysis During Cheese Manufacture and Ripening.” Journal of Dairy
Science V72 (6): pp. 1379–1400.
Fox, P. F., J. A. Lucey, and T. M. Cogan. 1990. “Glycolysis and Related Reactions During
Cheese Manufacture and Ripening.” Critical Reviews in Food Science and Nutrition
V29 (4): pp. 237–253.
Fox, P.F., and L. Stepaniak. 1993. “Enzymes in Cheese Technology.” International Dairy
Journal V3 (4–6): pp. 509–530.
Fox, P. F., and P. L. H. McSweeney. 1996. “Proteolysis in Cheese During Ripening.” Food
Reviews International V12 (4): pp. 457–509.
Fox, P. F., J. M. Wallace, S. Morgan, C. M. Lynch, E. J. Niland, and J. Tobin. 1996.
“Acceleration of Cheese Ripening.” Antonie Van Leeuwenhoek 70 (2): 271–297.
Fox, PF, PLH McSweeney, and CM Lynch. 1998. “Significance of Non-starter Lactic Acid
Bacteria in Cheddar Cheese.” Australian Journal of Dairy Technology V53 (2): pp. 7–
8.
Fox, P.F., and P.L.H. McSweeney 1998. “Milk Lipids.” In Dairy Chemistry and Biochemistry,
pp. 67–141. Springer.
Fox, P. F., Law J, P. L. H. McSweeney, and Wallace J.M. 1999. “Biochemistry of Cheese
Ripening.” In Cheese: Chemistry, Physics and Microbiology, pp. 382–450. Springer.
Fox, P. F., T.P. Guinee, T. M. Cogan, and P. L. H. McSweeney. 2000. “Enzymatic
Coagulation of Milk.” In Fundamentals of Cheese Science, pp. 98135. Springer.
Fox, P.F., and P.L.H. McSweeney. 2004. “Cheese: An Overview.” In Cheese Chemistry,
Physics and Microbiology, Volume 1:1–18. Academic Press.
202
Gao, Song, Ed S. Mooberry, and James L. Steele. 1998. “Use of 13C Nuclear Magnetic
Resonance and Gas Chromatography To Examine Methionine Catabolism by
Lactococci.” Applied and Environmental Microbiology V64 (12): pp. 4670–4675.
Gomez, Maria J., Sonia Garde, Pilar Gaya, Margarita Medina, and Manuel Nuñez. 1997.
“Relationship Between Level of Hydrophobic Peptides and Bitterness in Cheese Made
from Pasteurized and Raw Milk.” Journal of Dairy Research V64 (2): pp. 289–297.
Grandison, Alistair S., and Graeme D. Ford. 1986. “Effects of Variations in Somatic Cell
Count on the Rennet Coagulation Properties of Milk and on the Yield, Composition
and Quality of Cheddar Cheese.” Journal of Dairy Research V53 (4): pp. 645–655.
Grappin, Rémy, and Eric Beuvier. 1997. “Possible Implications of Milk Pasteurization on the
Manufacture and Sensory Quality of Ripened Cheese.” International Dairy Journal
V7 (12): pp. 751–761.
Green, Margaret L., and Donald J. Manning. 1982. “Development of Texture and Flavour in
Cheese and Other Fermented Products.” Journal of Dairy Research V49 (4): pp. 737–
748.
Grieve, Paul A., and Barry J. Kitchen. 1985. “Proteolysis in Milk: The Significance of
Proteinases Originating from Milk Leucocytes and a Comparison of the Action of
Leucocyte, Bacterial and Natural Milk Proteinases on Casein.” Journal of Dairy
Research 52 (1): 101–112.
Griffiths, M. W. 1986. “Use of milk enzymes as indices of heat treatment.” Journal of Food
Protection V49 (9): pp. 696–705.
Gripon, Jean-Claude, V Monnet, G Lambert, and M. J. Desmazeaud. 1991. “Microbial
enzymes in cheese ripening.” In Food enzymology, vol 1.
Guinee, T P. 2004. “Salting and the Role of Salt in Cheese.” International Journal of Dairy
Technology V57 (2-3): pp. 99–109.
Habibi‐Najafi, Mohammad B., Byong H. Lee, and Barry Law. 1996. “Bitterness in Cheese: A
Review.” Critical Reviews in Food Science and Nutrition V36 (5): pp. 397–411.
Heegaard, Christian W., Lone K. Rasmussen, and Peter A. Andreasen. 1994. “The
Plasminogen Activation System in Bovine Milk: Differential Localization of Tissuetype Plasminogen Activator and Urokinase in Milk Fractions Is Caused by Binding to
Casein and Urokinase Receptor.” Biochimica Et Biophysica Acta (BBA) - Molecular
Cell Research V1222 (1): pp. 45–55.
Hickey, D.K., K.N. Kilcawley, T.P. Beresford, and M.G. Wilkinson. 2007. “Lipolysis in
Cheddar Cheese Made from Raw, Thermized, and Pasteurized Milks.” Journal of
Dairy Science V90 (1): pp. 47–56.
Holland, Ross, and Tim Coolbear. 1996. “Purification of Tributyrin Esterase from
Lactococcus Lactis Subsp. Cremoris E8.” Journal of Dairy Research V63 (1): pp.
131–140.
Home, D.S., and J.M. Banks. 2004. “Rennet-induced Coagulation of Milk.” In Cheese
Chemistry, Physics and Microbiology, Volume 1:pp. 47–70. Academic Press.
Huffman, L. M.;, and Kristoffersen. 1984. “Role of lactose in Cheddar cheese manufacturing
and ripening.” New Zealand Journal of Dairy Science and Technology V19 (2): pp.
151–162.
Hurley, M.J, B.M O’Driscoll, A.L Kelly, and P.L.H McSweeney. 1999. “Novel Assay for the
Determination of Residual Coagulant Activity in Cheese.” International Dairy
Journal V9(8): pp 553–558.
203
Hutton, J.T., and S. Patton. 1952. “The Origin of Sulfhydryl Groups in Milk Proteins and
Their Contributions to ‘Cooked’ Flavor.” Journal of Dairy Science V35 (8): pp. 699–
705.
Jolly, Ramesh C., and Frank V. Kosikowski. 1975. “Quantification of Lactones in Ripening
Pasteurized Milk Blue Cheese Containing Added Microbial Lipases.” J. Agric. Food
Chem. V23 (6): pp. 1175–1176.
De Jong, Catrienus, and Herman T. Badings. 1990. “Determination of Free Fatty Acids in
Milk and Cheese Procedures for Extraction, Clean up, and Capillary Gas
Chromatographic Analysis.” Journal of High Resolution Chromatography V13 (2):
pp. 94–98.
K. Arfi, H.Spinnler, R. Tache, and P. bonnarme. 2002. “Production of Volatile Compounds by
Cheese-ripening Yeasts: Requirement for a Methanethiol Donor for S-methyl
Thioacetate Synthesis by Kluyveromyces Lactis.” Applied Microbiology and
Biotechnology V58 (4): pp. 503–510.
Knecht, E. 1999. “Lysomomal Cysteine Proteinases (2nd Edition).” Cell Biology International
V23 (8): pp 590–591.
Laan, Harry, Saw Eng Tan, Paul Bruinenberg, Gaëtan Limsowtin, and Malcolm Broome.
1998. “Aminopeptidase Activities of Starter and Non-Starter Lactic Acid Bacteria
Under Simulated Cheddar Cheese Ripening Conditions.” International Dairy Journal
V8 (4): pp. 267–274.
Lane, C.N., and P.F. Fox. 1996a. “Contribution of Starter and Adjunct Lactobacilli to
Proteolysis in Cheddar Cheese During Ripening.” International Dairy Journal 6 (7):
715–728.
Lane, Cáit N., and Patrick F. Fox. 1997. “Role of Starter Enzymes During Ripening of
Cheddar Cheese Made from Pasteurized Milk Under Controlled Microbiological
Conditions.” International Dairy Journal V7 (1): pp. 55–63.
Larsen, Lotte B., Connie Benfeldt, Lone K. Rasmussen, and Torben E. Petersen. 1996.
“Bovine Milk Procathepsin D and Cathepsin D: Coagulation and Milk Protein
Degradation.” Journal of Dairy Research V63 (1): pp 119–130.
Lau, Kum Y., David M. Barbano, and Robert R. Rasmussen. 1990. “Influence of
Pasteurization on Fat and Nitrogen Recoveries and Cheddar Cheese Yield.” Journal of
Dairy Science V73 (3): pp. 561–570.
Lau, Kum Y. Lau, David M. Barbano, and Robert R. Rasmussen. 1991. “Influence of
Pasteurization of Milk on Protein Breakdown in Cheddar Cheese During Aging.”
Journal of Dairy Science V74 (3): pp. 727–740.
Law, A. J. R., J. M. Banks, D. S. Horne, J. Leaver, and I. G. West. 1994. “Denaturation of the
whey proteins in heated milk and their incorporation into Cheddar cheese.”
Milchwissenschaft V49 (2): pp. 63–67.
Lee, Won-Jae, Dattatreya S. Banavara, Joanne E. Hughes, Jason K. Christiansen, James L.
Steele, Jeffery R. Broadbent, and Scott A. Rankin. 2007. “Role of Cystathionine βLyase in Catabolism of Amino Acids to Sulfur Volatiles by Genetic Variants of
Lactobacillus Helveticus CNRZ 32.” Applied and Environmental Microbiology V73
(9): pp. 3034–3039.
Lin, J. C. C., and I. J. Jeon. 1985. “Headspace Gas Sampling/GC Method for the Quantitative
Analysis of Volatile Compounds in Cheese.” Journal of Food Science V50 (3): pp.
843–844.
204
Liu, Shao-Quan, Ross Holland, and Vaughan L Crow. 2001. “Purification and Properties of
Intracellular Esterases from Streptococcus Thermophilus.” International Dairy
Journal V11 (1–2): pp. 27–35.
Lloyd, G. T., J. F. Horwood, and I. Barlow. 1980. “The effect of yoghurt culture YB on the
flavour and maturation of Cheddar cheese.” Australian Journal of Dairy Technology
V35 (4): pp. 137–139.
Lynch, C.M., P.L.H. McSweeney, P.F. Fox, T.M. Cogan, and F.D. Drinan. 1996.
“Manufacture of Cheddar Cheese with and Without Adjunct Lactobacilli Under
Controlled Microbiological Conditions.” International Dairy Journal V6 (8–9): pp.
851–867.
Lynch, C.M., D.D. Muir, J.M. Banks, P.L.H. McSweeney, and P.F. Fox. 1999. “Influence of
Adjunct Cultures of Lactobacillus Paracasei Ssp. Paracasei or Lactobacillus Plantarum
on Cheddar Cheese Ripening.” Journal of Dairy Science V82 (8): pp. 1618–1628.
Lynch, Joanna M., David M. Barbano, and J. Richard Fleming. 2002. “Determination of the
Total Nitrogen Content of Hard, Semihard, and Processed Cheese by the Kjeldahl
Method: Collaborative Study.” Journal of AOAC International V85 (2): pp. 445–455.
Madkor, S.A., P.S. Tong, and M. El Soda. 2000. “Ripening of Cheddar Cheese with Added
Attenuated Adjunct Cultures of Lactobacilli.” Journal of Dairy Science V83 (8): pp.
1684–1691.
Manning, D. J., Helen R. Chapman, and Zena D. Hosking. 1976. “The Production of Sulphur
Compounds in Cheddar Cheese and Their Significance in Flavour Development.”
Journal of Dairy Research V43 (2): pp. 313–320.
Manning, Donald J., and Carolyn Moore. 1979. “Headspace Analysis of Hard Cheeses.”
Journal of Dairy Research V46 (3): pp. 539–545.
McSweeney, P.L.H., P.F. Fox, J.A. Lucey, K.N. Jordan, and T.M. Cogan. 1993. “Contribution
of the Indigenous Microflora to the Maturation of Cheddar Cheese.” International
Dairy Journal V3 (7): pp. 613–634.
McSweeney, Paul L. H., Sylvie Pochet, Patrick F. Fox, and Aine Healy. 1994. “Partial
Identification of Peptides from the Water-Insoluble Fraction of Cheddar Cheese.”
Journal of Dairy Research V61 (4): pp. 587–590.
McSweeney, P.L.H., P.F. Fox, and N.F. Olson. 1995. “Proteolysis of Bovine Caseins by
Cathepsin D: Preliminary Observations and Comparison with Chymosin.”
International Dairy Journal V5 (4): pp. 321–336.
McSweeney, Paul L.H., and Maria José Sousa. 2000. “Biochemical Pathways for the
Production of Flavour Compounds in Cheeses During Ripening: A Review.” Le Lait
V80 (3): pp. 293–324..
Metwalli, Ali A.M, Harmen H.J de Jongh, and Martinus A.J.S van Boekel. 1998. “Heat
Inactivation of Bovine Plasmin.” International Dairy Journal V8 (1): pp. 47–56.
Milo, C., and G. A. Reineccius. 1997. “Identification and Quantification of Potent Odorants in
Regular-Fat and Low-Fat Mild Cheddar Cheese.” Journal of Agricultural and Food
Chemistry V45 (9): pp. 3590–3594.
Mistry, V.V., L.E. Metzger, and J.L. Maubois. 1996. “Use of Ultrafiltered Sweet Buttermilk in
the Manufacture of Reduced Fat Cheddar Cheese.” Journal of Dairy Science V79 (7):
pp. 1137–1145.
Mistry, V.V., and J.-L. Maubois. 2004. “Application of Membrane Separation Technology to
Cheese Production.” In Cheese Chemistry, Physics and Microbiology, Volume 1:pp.
261–VII. Academic Press.
205
Moio, Luigi, Dominique Langlois, Patrick Etievant, and Francesco Addeo. 1993. “Powerful
Odorants in Bovine, Ovine, Caprine and Water Buffalo Milk Determined by Means of
Gas Chromatography–olfactometry.” Journal of Dairy Research V60 (2): pp. 215–
222.
Molimard, P., and H.E. Spinnler. 1996. “Review: Compounds Involved in the Flavor of
Surface Mold-Ripened Cheeses: Origins and Properties.” Journal of Dairy Science
V79 (2): pp. 169–184.
Mooney, J.S, P.F Fox, A Healy, and J Leaver. 1998. “Identification of the Principal Waterinsoluble Peptides in Cheddar Cheese.” International Dairy Journal V8 (9): pp. 813–
818.
Muir, D.D., J.M. Banks, and E.A. Hunter. 1996. “Sensory Properties of Cheddar Cheese:
Effect of Starter Type and Adjunct.” International Dairy Journal V6 (4): pp. 407–
423.
Mulvihill, Donald M., and Patrick F. Fox. 1979. “Proteolytic Specificity of Chymosin on
Bovine Αs1-Casein.” Journal of Dairy Research V46 (4): pp. 641–651.
Naughton, M. A., and F. Sanger. 1961. “Purification and Specificity of Pancreatic Elastase.”
Biochemical Journal V78 (1): pp. 156–163.
Niven, C. F., L. G. Buettner, and James B. Evans. 1954. “Thermal Tolerance Studies on the
Heterofermentative Lactobacilli That Cause Greening of Cured Meat Products.”
Applied Microbiology V2 (1): pp. 26–29.
O’Keeffe, R. B., P. F. Fox, and C. Daly. 1976. “Contribution of Rennet and Starter Proteases
to Proteolysis in Cheddar Cheese.” Journal of Dairy Research 43 (1): 97–107.
O’Keeffe, Arthur M., Patrick F. Fox, and Charles Daly. 1978. “Proteolysis in Cheddar Cheese:
Role of Coagulant and Starter Bacteria.” Journal of Dairy Research V45 (3): pp. 465–
477.
Olivecrona, T., S. Vilaró, G. Bengtsson-Olivecrona, and P. F Fox. 1992. “Indigenous enzyme
in milk. II. Lipases in milk.” In Advanced Dairy Chemistry 1,pp 292–310..
Payne, Martin J., and Rohan G. Kroll. 1991. “Methods for the Separation and Concentration
of Bacteria from Foods.” Trends in Food Science & Technology V2 (0): pp. 315–319.
Peterson, S.D., and R.T. Marshall. 1990. “Nonstarter Lactobacilli in Cheddar Cheese: A
Review.” Journal of Dairy Science V73 (6): pp. 1395–1410.
Piraino, Paolo, Eugenio Parente, and Paul L. H. McSweeney. 2004. “Processing of
Chromatographic Data for Chemometric Analysis of Peptide Profiles from Cheese
Extracts:  A Novel Approach.” Journal of Agricultural and Food Chemistry 52 (23):
6904–691.
Precetti, A.S., M.P. Oria, and S.S. Nielsen. 1997. “Presence in Bovine Milk of Two Protease
Inhibitors of the Plasmin System.” Journal of Dairy Science V80 (8): pp. 1490–1496.
Preininger, Martin, and Werner Grosch. 1994. “Evaluation of Key Odorants of the Neutral
Volatiles of Emmentaler Cheese by the Calculation of Odour Activity Values.” LWT Food Science and Technology V27 (3): pp. 237–244.
Pripp, A.H, Shakeel-Ur-Rehman, P.L.H McSweeney, and P.F Fox. 1999. “Multivariate
Statistical Analysis of Peptide Profiles and Free Amino Acids to Evaluate Effects of
Single-strain Starters on Proteolysis in Miniature Cheddar-type Cheeses.”
International Dairy Journal V9 (7): pp. 473–479.
Pripp, Are Hugo, Leszek Stepaniak, and Terje Sørhaug. 2000. “Chemometrical Analysis of
Proteolytic Profiles During Cheese Ripening.” International Dairy Journal V10 (4):
pp. 249–253.
206
Puchades, R., L. Lemieux, and R.e. Simard. 1989. “Evolution of Free Amino Acids During
the Ripening of Cheddar Cheese Containing Added Lactobacilli Strains.” Journal of
Food Science V54 (4): pp. 885–888.
Qian, M., and G. Reineccius. 2002. “Identification of Aroma Compounds in ParmigianoReggiano Cheese by Gas Chromatography/Olfactometry.” Journal of Dairy Science
V85 (6): pp. 1362–1369.
Rehman, Shakeel-Ur, J.M Banks, E.Y Brechany, D.D Muir, P.L.H McSweeney, and P.F Fox.
2000. “Influence of Ripening Temperature on the Volatiles Profile and Flavour of
Cheddar Cheese Made from Raw or Pasteurised Milk.” International Dairy Journal
V10 (1-2): pp. 55–65.
Richardson, B. C., L. K. Creamer, K. N. Pearce, and R. E. Munford. 1974. “Comparative
Micelle Structure: II. Structure and Composition of Casein Micelles in Ovine and
Caprine Milk as Compared with Those in Bovine Milk.” Journal of Dairy Research
V41 (2): pp. 239–247.
Roy, Denis, Isabelle Mainville, and Francine Mondou. 1997. “Selective Enumeration and
Survival of Bifidobacteria in Fresh Cheese.” International Dairy Journal V7 (12): pp.
785–793.
Sablé, Sophie, and Gilles Cottenceau. 1999. “Current Knowledge of Soft Cheeses Flavor and
Related Compounds.” J. Agric. Food Chem. V47 (12): pp. 4825–4836.
Seefeldt, K.E., and B.C. Weimer. 2000. “Diversity of Sulfur Compound Production in Lactic
Acid Bacteria.” Journal of Dairy Science V83 (12): pp. 2740–2746.
Shakeel-Ur-Rehman, J. M Banks, P. L. H Mcsweeney, and P. F Fox. 2000. “Effect of ripening
temperature on the growth and significance of non-starter lactic acid bacteria in
Cheddar cheese made from raw or pasteurised milk.” International dairy journal V10
(1-2): pp. 45–53.
Singh, T.K., P.F. Fox, P. Højrup, and A. Healy. 1994. “A Scheme for the Fractionation of
Cheese Nitrogen and Identification of Principal Peptides.” International Dairy
Journal V4 (2): pp. 111–122.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1995. “Water-Soluble Peptides in Cheddar
Cheese: Isolation and Identification of Peptides in the Diafiltration Retentate of the
Water-Soluble Fraction.” Journal of Dairy Research V62 (4). 629–640.
Singh, Tanoj K., Patrick F. Fox, and Áine Healy. 1997. “Isolation and Identification of Further
Peptides in the Diafiltration Retentate of the Water-soluble Fraction of Cheddar
Cheese.” Journal of Dairy Research V64 (3): pp. 433–443.
Singh, T. K, M. A Drake, and K. R Cadwallader. 2003. “Flavor of Cheddar Cheese: A
Chemical and Sensory Perspective.” Comprehensive Reviews in Food Science and
Food Safety V2 (4): pp. 166–189.
Smacchi, Emanuele, and Marco Gobbetti. 1998. “Purification and Characterization of
Cystathionine Γ-lyase from Lactobacillus Fermentum DT41.” FEMS Microbiology
Letters V166 (2): pp. 197–202.
Smit, Gerrit, Bart A. Smit, and Wim J.M. Engels. 2005. “Flavour Formation by Lactic Acid
Bacteria and Biochemical Flavour Profiling of Cheese Products.” FEMS Microbiology
Reviews V29 (3): pp. 591–610.
Smith, A. M., and S. Nakai. 1990. “Classification of cheese varieties by multivariate analysis
of HPLC profiles.” Canadian Institute of Food Science and Technology Journal V23
(1): pp. 53–58.
Sousa, M.J, Y Ardö, and P.L.H McSweeney. 2001. “Advances in the Study of Proteolysis
During Cheese Ripening.” International Dairy Journal V11 (4-7): pp. 327–345.
207
Stuart, Mark R., Lan Szu Chou, and Bart C. Weimer. 1999. “Influence of Carbohydrate
Starvation and Arginine on Culturability and Amino Acid Utilization of Lactococcus
Lactis Subsp. Lactis.” Applied and Environmental Microbiology V65 (2): pp. 665–
673.
Subramanian, A., W.J. Harper, and L.E. Rodriguez-Saona. 2009. “Cheddar Cheese
Classification Based on Flavor Quality Using a Novel Extraction Method and Fourier
Transform Infrared Spectroscopy.” Journal of Dairy Science V92 (1): pp. 87–94.
Suriyaphan, Orasa, MaryAnne Drake, X. Q. Chen, and Keith R. Cadwallader. 2001.
“Characteristic Aroma Components of British Farmhouse Cheddar Cheese.” Journal
of Agricultural and Food Chemistry V49 (3): pp. 1382–1387.
Tammam, J. D, A. G Williams, J. Noble, D. Lloyd, J. D Tammam, A. G Williams, J. Noble,
and D. Lloyd. 2000. “Amino acid fermentation in non‐starter Lactobacillus spp.
isolated from Cheddar cheese, Amino acid fermentation in non‐starter Lactobacillus
spp. isolated from Cheddar cheese.” Letters in Applied Microbiology, Letters in
Applied Microbiology V30 (5): pp. 370–374.
Tanaka, Hidehiko, Nobuyoshi Esaki, and Kenji Soda. 1985. “A Versatile Bacterial Enzyme:
L-methionine Γ-lyase.” Enzyme and Microbial Technology V7 (11): pp. 530–537.
Thomsen, D., and H. Stapelfeldt. 1990. “Yield and quality of Danbo 45+ cheese from milk
heat treated at temperatures between 72 and 95°C.” V44: pp. 39.
Tsakalidou, E., and G. Kalantzopoulos. 1992. “Purification and Partial Characterization of an
Esterase from Lactococcus Lactis Ssp Lactis Strain ACA-DC 127.” Le Lait V72 (6):
pp. 533–543.
Turner, K. W, and T. D Thomas. 1980. “Lactose fermentation in Cheddar cheese and the
effect of salt.” New Zealand Journal of Dairy Science and Technology V15 (3): pp.
265–276.
Upadhyay, V.K., P.L.H. McSweeney, A.A.A. Magboul, and P.F. Fox. 2004. “Proteolysis in
Cheese During Ripening.” In Cheese Chemistry, Physics and Microbiology, Volume
1:pp. 391–433. Academic Press.
Urbach, Gerda. 1993. “Relations Between Cheese Flavour and Chemical Composition.”
International Dairy Journal V3 (4–6): pp. 389–422.
Urbach, Gerda. 1995. “Contribution of Lactic Acid Bacteria to Flavour Compound Formation
in Dairy Products.” International Dairy Journal V5 (8): pp. 877–903.
USDA. 2011. Dairy Products Annual Sumary.
Vazquez-Landaverde, P.A., J.A. Torres, and M.C. Qian. 2006. “Quantification of Trace
Volatile Sulfur Compounds in Milk by Solid-Phase Microextraction and Gas
Chromatography–Pulsed Flame Photometric Detection.” Journal of Dairy Science
V89 (8): pp. 2919–2927.
Verdi, Robert J., and David M. Barbano. 1991. “Properties of Proteases from Milk Somatic
Cells and Blood Leukocytes.” Journal of Dairy Science V74 (7): pp. 2077–2081.
Visser, Servas 1983. “Bitter flavour in cheese, 2: model studies on the formation and
degradation of bitter peptides by proteolytic enzymes from calf rennet, starter cells
and starter cell fractions.” Netherlands Milk and Dairy Journal V37 (3): pp. 169–180.
Visser, Servaas. 1993. “Proteolytic Enzymes and Their Relation to Cheese Ripening and
Flavor: An Overview.” Journal of Dairy Science V76 (1): pp. 329–350
Voigt, Daniela D., François Chevalier, John A. Donaghy, Margaret F. Patterson, Michael C.
Qian, and Alan L. Kelly. 2012. “Effect of High-pressure Treatment of Milk for
Cheese Manufacture on Proteolysis, Lipolysis, Texture and Functionality of Cheddar
208
Cheese During Ripening.” Innovative Food Science & Emerging Technologies V13
(0): pp. 23–30.
Wallace J.M., and Fox P.F. 1997. “Effect of Adding Free Amino Acids to Cheddar Cheese
Curd on Proteolysis, Flavour and Texture Development.” International Dairy Journal
V7 (2): pp. 157–167.
Walstra, Pieter, T. J Geurts, A. Noomen, A. Jellema, and M. A. J. S. van Boekel. 1999.
“Cheese Varieties.” In Dairy technology: principles of milk properties and processes,
pp 651–708. Marcel Dekker.
Walstra, Pieter, Jan T. M. Wouters, and Tom J. Geurts. 2006. Dairy Science And Technology.
CRC Press.
Weimer, B. C. 2007. Improving the flavour of cheese. CRC Press.
Weimer, Bart, Kimberly Seefeldt, and Benjamin Dias. 1999. “Sulfur Metabolism in Bacteria
Associated with Cheese.” Antonie Van Leeuwenhoek 76 (1): 247–261.
Wijesundera, C., and L. Drury. “Role of milk fat in production of cheddar cheese flavour
using a fat-substituted cheese model.” Australian journal of dairy technology V54 (1):
pp. 28–35.
Wilkinson, Martin G., Timothy P. Guinee, Daniel M. O’Callaghan, and Patrick F. Fox. 1994.
“Autolysis and Proteolysis in Different Strains of Starter Bacteria During Cheddar
Cheese Ripening.” Journal of Dairy Research V61 (2): pp. 249–262.
De Wit, M., G. Osthoff, B.C. Viljoen, and A. Hugo. 2005. “A Comparative Study of Lipolysis
and Proteolysis in Cheddar Cheese and Yeast-inoculated Cheddar Cheeses During
Ripening.” Enzyme and Microbial Technology V37 (6): pp. 606–616.
de Wit, Rudy, and Hans Nieuwenhuijse. 2008. “Kinetic Modelling of the Formation of
Sulphur-containing Flavour Components During Heat-treatment of Milk.”
International Dairy Journal V18 (5): pp. 539–547.
Wong, N. P, R. Ellis, and D. E LaCroix. “Quantitative determination of lactones in Cheddar
cheese.” Journal of Dairy Science V58: pp. 1437–1441.
Yvon, Mireille, Claire Chabanet, and Jean-Pierre Pélissier. 1989. “Solubility of Peptides in
Trichloroacetic Acid (TCA) Solutions Hypothesis on the Precipitation Mechanism.”
International Journal of Peptide and Protein Research V34 (3): pp. 166–176.
Yvon, Mireille, and Liesbeth Rijnen. 2001. “Cheese Flavour Formation by Amino Acid
Catabolism.” International Dairy Journal V11 (4–7): pp. 185–201.
Zehentbauer, Gerhard, and G. A. Reineccius. 2002. “Determination of Key Aroma
Components of Cheddar Cheese Using Dynamic Headspace Dilution Assay.” Flavour
and Fragrance Journal V17 (4): pp. 300–305.
209
APPENDICES
APENDIX A
ANOVAS
FFA
Heat-shocked Vs Pasteurized
Table 2 Anova FFA of Heat-shocked Vs Pasteurized
General Linear Model: C4:0, C6:0, ... versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
7
Values
HS, P
10M, 12M, 14M, 2M, 4M, 6M, 8M
Analysis of Variance for C4:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
8.026
605.649
3.354
*
617.029
Adj SS
8.026
605.649
3.354
*
Adj MS
8.026
100.941
0.559
*
F
14.36
180.56
**
P
0.009
0.000
Analysis of Variance for C6:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
2.9257
122.0326
1.5343
*
126.4926
Adj SS
2.9257
122.0326
1.5343
*
Adj MS
2.9257
20.3388
0.2557
*
F
11.44
79.54
**
P
0.015
0.000
Analysis of Variance for C8:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
0.6487
19.9431
0.3181
*
20.9099
Adj SS
0.6487
19.9431
0.3181
*
Adj MS
0.6487
3.3239
0.0530
*
F
12.24
62.70
**
P
0.013
0.000
Analysis of Variance for C10:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
DF
1
6
6
Seq SS
3.1114
71.3699
0.8686
Adj SS
3.1114
71.3699
0.8686
Adj MS
3.1114
11.8950
0.1448
F
21.49
82.17
**
P
0.004
0.000
210
Error
Total
0
13
*
75.3499
*
*
Analysis of Variance for C12:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
3.2335
102.7589
1.8224
*
107.8147
Adj SS
3.2335
102.7589
1.8224
*
Adj MS
3.2335
17.1265
0.3037
*
F
10.65
56.39
**
P
0.017
0.000
Analysis of Variance for C14:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
96.68
1213.26
23.50
*
1333.44
Adj SS
96.68
1213.26
23.50
*
Adj MS
96.68
202.21
3.92
*
F
24.68
51.63
**
P
0.003
0.000
Analysis of Variance for C14:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
0.33913
8.01428
0.45428
*
8.80769
Adj SS
0.33913
8.01428
0.45428
*
Adj MS
0.33913
1.33571
0.07571
*
F
4.48
17.64
**
P
0.079
0.001
Analysis of Variance for C16:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
503.88
3988.00
150.99
*
4642.87
Adj SS
503.88
3988.00
150.99
*
Adj MS
503.88
664.67
25.17
*
F
20.02
26.41
**
P
0.004
0.000
Analysis of Variance for C16:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
0.6079
20.0450
0.7335
*
21.3863
Adj SS
0.6079
20.0450
0.7335
*
Adj MS
0.6079
3.3408
0.1223
*
F
4.97
27.33
**
P
0.067
0.000
Analysis of Variance for C18:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
25.42
1361.62
42.00
*
1429.04
Adj SS
25.42
1361.62
42.00
*
Adj MS
25.42
226.94
7.00
*
F
3.63
32.42
**
P
0.105
0.000
Analysis of Variance for C18:1, using Adjusted SS for Tests
211
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
92.14
6277.51
247.49
*
6617.13
Adj SS
92.14
6277.51
247.49
*
Adj MS
92.14
1046.25
41.25
*
F
2.23
25.36
**
P
0.186
0.001
Analysis of Variance for C18:2, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
0.024
1057.288
50.452
*
1107.764
Adj SS
0.024
1057.288
50.452
*
Adj MS
0.024
176.215
8.409
*
F
0.00
20.96
**
P
0.959
0.001
Analysis of Variance for C18:3, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
0.00072
2.59199
0.09388
*
2.68659
Adj SS
0.00072
2.59199
0.09388
*
Adj MS
0.00072
0.43200
0.01565
*
F
0.05
27.61
**
P
0.837
0.000
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
TCCA Vs CRP
Table 3 Anova FFA of TCCA Vs CRP
General Linear Model: C4:0, C6:0, ... versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
7
Values
CRP, TCCA
10M, 12M, 14M, 2M, 4M, 6M, 8M
Analysis of Variance for C4:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
26.612
444.280
40.857
*
511.749
Adj SS
26.612
444.280
40.857
*
Adj MS
26.612
74.047
6.810
*
F
3.91
10.87
**
P
0.095
0.005
Analysis of Variance for C6:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
5.9080
73.1742
5.4894
*
84.5717
Adj SS
5.9080
73.1742
5.4894
*
Adj MS
5.9080
12.1957
0.9149
*
F
6.46
13.33
**
P
0.044
0.003
212
Analysis of Variance for C8:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
7.9018
11.0297
0.7586
*
19.6901
Adj SS
7.9018
11.0297
0.7586
*
Adj MS
7.9018
1.8383
0.1264
*
F
62.50
14.54
**
P
0.000
0.002
Analysis of Variance for C10:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
47.0056
51.2884
10.9329
*
109.2269
Adj SS
47.0056
51.2884
10.9329
*
Adj MS
47.0056
8.5481
1.8221
*
F
25.80
4.69
**
P
0.002
0.041
Analysis of Variance for C12:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
80.376
65.265
9.070
*
154.711
Adj SS
80.376
65.265
9.070
*
Adj MS
80.376
10.878
1.512
*
F
53.17
7.20
**
P
0.000
0.015
Analysis of Variance for C14:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
817.27
498.97
167.39
*
1483.62
Adj SS
817.27
498.97
167.39
*
Adj MS
817.27
83.16
27.90
*
F
29.29
2.98
**
P
0.002
0.105
Analysis of Variance for C14:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
10.2390
5.6458
1.7850
*
17.6698
Adj SS
10.2390
5.6458
1.7850
*
Adj MS
10.2390
0.9410
0.2975
*
F
34.42
3.16
**
P
0.001
0.093
Analysis of Variance for C16:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
1152.33
1788.07
273.61
*
3214.01
Adj SS
1152.33
1788.07
273.61
*
Adj MS
1152.33
298.01
45.60
*
F
25.27
6.54
**
P
0.002
0.019
Analysis of Variance for C16:1, using Adjusted SS for Tests
Source
treatment
time
DF
1
6
Seq SS
14.9829
18.9973
Adj SS
14.9829
18.9973
Adj MS
14.9829
3.1662
F
22.34
4.72
P
0.003
0.040
213
treatment*time
Error
Total
6
0
13
4.0239
*
38.0041
4.0239
*
0.6707
*
**
Analysis of Variance for C18:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
296.820
537.598
34.560
*
868.978
Adj SS
296.820
537.598
34.560
*
Adj MS
296.820
89.600
5.760
*
F
51.53
15.56
**
P
0.000
0.002
Analysis of Variance for C18:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
3159.42
2227.26
613.90
*
6000.58
Adj SS
3159.42
2227.26
613.90
*
Adj MS
3159.42
371.21
102.32
*
F
30.88
3.63
**
P
0.001
0.071
Analysis of Variance for C18:2, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
3.4844
93.9442
7.2245
*
104.6531
Adj SS
3.4844
93.9442
7.2245
*
Adj MS
3.4844
15.6574
1.2041
*
F
2.89
13.00
**
P
0.140
0.003
Analysis of Variance for C18:3, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
6
6
0
13
Seq SS
12.2826
6.4034
3.5308
*
22.2169
Adj SS
12.2826
6.4034
3.5308
*
Adj MS
12.2826
1.0672
0.5885
*
F
20.87
1.81
**
P
0.004
0.244
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
Adjunct culture Vs Not
Table 4 Anova FFA of Adjunct culture Vs Not
General Linear Model: C4:0, C6:0, ... versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
5
Values
A, N
2, 4, 6, 8, 10
214
Analysis of Variance for C4:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
12.877
352.397
1.314
*
366.588
Adj SS
12.877
352.397
1.314
*
Adj MS
12.877
88.099
0.329
*
F
39.19
268.16
**
P
0.003
0.000
Analysis of Variance for C6:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
3.9297
44.7948
0.3472
*
49.0717
Adj SS
3.9297
44.7948
0.3472
*
Adj MS
3.9297
11.1987
0.0868
*
F
45.28
129.03
**
P
0.003
0.000
Analysis of Variance for C8:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
0.90706
6.38731
0.09790
*
7.39227
Adj SS
0.90706
6.38731
0.09790
*
Adj MS
0.90706
1.59683
0.02448
*
F
37.06
65.24
**
P
0.004
0.001
Analysis of Variance for C10:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
14.2226
28.8278
0.5657
*
43.6162
Adj SS
14.2226
28.8278
0.5657
*
Adj MS
14.2226
7.2069
0.1414
*
F
100.56
50.96
**
P
0.001
0.001
Analysis of Variance for C12:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
21.8046
37.0475
1.7017
*
60.5538
Adj SS
21.8046
37.0475
1.7017
*
Adj MS
21.8046
9.2619
0.4254
*
F
51.25
21.77
**
P
0.002
0.006
Analysis of Variance for C14:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
64.010
441.849
5.952
*
511.811
Adj SS
64.010
441.849
5.952
*
Adj MS
64.010
110.462
1.488
*
F
43.02
74.24
**
P
0.003
0.001
Analysis of Variance for C14:1, using Adjusted SS for Tests
Source
treatment
DF
1
Seq SS
0.01894
Adj SS
0.01894
Adj MS
0.01894
F
0.95
P
0.386
215
time
treatment*time
Error
Total
4
4
0
9
3.44150
0.08004
*
3.54048
3.44150
0.08004
*
0.86038
0.02001
*
43.00
**
0.002
Analysis of Variance for C16:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
131.04
2371.44
28.37
*
2530.86
Adj SS
131.04
2371.44
28.37
*
Adj MS
131.04
592.86
7.09
*
F
18.48
83.59
**
P
0.013
0.000
Analysis of Variance for C16:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
3.8062
18.6749
0.1284
*
22.6095
Adj SS
3.8062
18.6749
0.1284
*
Adj MS
3.8062
4.6687
0.0321
*
F
118.54
145.41
**
P
0.000
0.000
Analysis of Variance for C18:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
45.274
299.232
6.749
*
351.256
Adj SS
45.274
299.232
6.749
*
Adj MS
45.274
74.808
1.687
*
F
26.83
44.33
**
P
0.007
0.001
Analysis of Variance for C18:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
183.67
2647.99
51.89
*
2883.55
Adj SS
183.67
2647.99
51.89
*
Adj MS
183.67
662.00
12.97
*
F
14.16
51.03
**
P
0.020
0.001
Analysis of Variance for C18:2, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
7.060
82.232
7.758
*
97.049
Adj SS
7.060
82.232
7.758
*
Adj MS
7.060
20.558
1.939
*
F
3.64
10.60
**
P
0.129
0.021
Analysis of Variance for C18:3, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
0.28568
1.18875
0.15069
*
1.62512
Adj SS
0.28568
1.18875
0.15069
*
Adj MS
0.28568
0.29719
0.03767
*
** Denominator of F-test is zero or undefined.
F
7.58
7.89
**
P
0.051
0.035
216
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
Good Vs Week
Table 5 Anova FFA of Good Vs Weak
General Linear Model: C4:0, C6:0, ... versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
5
Values
Good, Weak
2, 6, 9, 12, 15
Analysis of Variance for C4:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
16.23
899.80
60.72
*
976.75
Adj SS
16.23
899.80
60.72
*
Adj MS
16.23
224.95
15.18
*
F
1.07
14.82
**
P
0.359
0.011
Analysis of Variance for C6:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
18.901
155.502
12.161
*
186.564
Adj SS
18.901
155.502
12.161
*
Adj MS
18.901
38.876
3.040
*
F
6.22
12.79
**
P
0.067
0.015
Analysis of Variance for C8:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
31.7506
26.5553
3.2097
*
61.5156
Adj SS
31.7506
26.5553
3.2097
*
Adj MS
31.7506
6.6388
0.8024
*
F
39.57
8.27
**
P
0.003
0.032
Analysis of Variance for C10:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
15.901
87.759
12.757
*
116.417
Adj SS
15.901
87.759
12.757
*
Adj MS
15.901
21.940
3.189
*
F
4.99
6.88
**
P
0.089
0.044
Analysis of Variance for C12:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
DF
1
4
4
0
Seq SS
20.336
135.429
31.303
*
Adj SS
20.336
135.429
31.303
*
Adj MS
20.336
33.857
7.826
*
F
2.60
4.33
**
P
0.182
0.093
217
Total
9
187.068
Analysis of Variance for C14:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
147.99
1152.15
107.94
*
1408.08
Adj SS
147.99
1152.15
107.94
*
Adj MS
147.99
288.04
26.99
*
F
5.48
10.67
**
P
0.079
0.021
Analysis of Variance for C14:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
0.7794
10.9725
1.5122
*
13.2641
Adj SS
0.7794
10.9725
1.5122
*
Adj MS
0.7794
2.7431
0.3781
*
F
2.06
7.26
**
P
0.224
0.040
Analysis of Variance for C16:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
2758.6
6210.3
347.2
*
9316.2
Adj SS
2758.6
6210.3
347.2
*
Adj MS
2758.6
1552.6
86.8
*
F
31.78
17.88
**
P
0.005
0.008
Analysis of Variance for C16:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
86.795
51.875
7.414
*
146.084
Adj SS
86.795
51.875
7.414
*
Adj MS
86.795
12.969
1.853
*
F
46.83
7.00
**
P
0.002
0.043
Analysis of Variance for C18:0, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
22.56
2848.10
460.17
*
3330.83
Adj SS
22.56
2848.10
460.17
*
Adj MS
22.56
712.02
115.04
*
F
0.20
6.19
**
P
0.681
0.053
Analysis of Variance for C18:1, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
2415.7
11093.8
871.1
*
14380.6
Adj SS
2415.7
11093.8
871.1
*
Adj MS
2415.7
2773.4
217.8
*
F
11.09
12.73
**
P
0.029
0.015
Analysis of Variance for C18:2, using Adjusted SS for Tests
Source
treatment
DF
1
Seq SS
859.53
Adj SS
859.53
Adj MS
859.53
F
13.92
P
0.020
218
time
treatment*time
Error
Total
4
4
0
9
346.00
246.92
*
1452.45
346.00
246.92
*
86.50
61.73
*
1.40
**
0.376
Analysis of Variance for C18:3, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
48.3241
6.8976
6.6459
*
61.8677
Adj SS
48.3241
6.8976
6.6459
*
Adj MS
48.3241
1.7244
1.6615
*
F
29.09
1.04
**
P
0.006
0.486
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
VSC’s
Heat-shocked Vs Pasteurized
Table 6 Anova VSC of Heat-Shocked Vs Pasteurized
General Linear Model: DMS, H2S, MeSH versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
10
Values
HS, P
2, 5, 6, 7, 8, 9, 10, 12, 14, 15
Analysis of Variance for DMS, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
9
9
0
19
Seq SS
2108.64
1316.67
295.56
*
3720.87
Adj SS
2108.64
1316.67
295.56
*
Adj MS
2108.64
146.30
32.84
*
F
64.21
4.45
**
P
0.000
0.018
Analysis of Variance for H2S, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
9
9
0
19
Seq SS
291.20
1245.71
1245.30
*
2782.21
Adj SS
291.20
1245.71
1245.30
*
Adj MS
291.20
138.41
138.37
*
F
2.10
1.00
**
P
0.181
0.500
Analysis of Variance for MeSH, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
9
9
0
19
Seq SS
0.189718
0.596347
0.137296
*
0.923361
Adj SS
0.189718
0.596347
0.137296
*
Adj MS
0.189718
0.066261
0.015255
*
F
12.44
4.34
**
P
0.006
0.020
219
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
TCCA Vs CRP
Table 7 Anova VSC TCCA Vs CRP
General Linear Model: DMS, H2S, MeSH versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
8
Values
CRP, TCCA
10M, 12M, 14M, 15M, 2M, 4M, 6M, 8M
Analysis of Variance for DMS, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
7
7
0
15
Seq SS
10.461
743.070
16.942
*
770.472
Adj SS
10.461
743.070
16.942
*
Adj MS
10.461
106.153
2.420
*
F
4.32
43.86
**
P
0.076
0.000
Analysis of Variance for H2S, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
7
7
0
15
Seq SS
76.434
147.463
84.326
*
308.223
Adj SS
76.434
147.463
84.326
*
Adj MS
76.434
21.066
12.047
*
F
6.34
1.75
**
P
0.040
0.239
Analysis of Variance for MeSH, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
7
7
0
15
Seq SS
0.009245
0.170565
0.022100
*
0.201909
Adj SS
0.009245
0.170565
0.022100
*
Adj MS
0.009245
0.024366
0.003157
*
F
2.93
7.72
**
P
0.131
0.008
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
Adjunct culture Vs Not
Table 8 Anova of VSC Adjunct Vs Not
General Linear Model: DMS, H2S, MeSH versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
5
Values
adj, No
10M, 2M, 4M, 6M, 8M
220
Analysis of Variance for DMS, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
24.643
432.056
6.056
*
462.755
Adj SS
24.643
432.056
6.056
*
Adj MS
24.643
108.014
1.514
*
F
16.28
71.34
**
P
0.016
0.001
Analysis of Variance for H2S, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
116.266
87.452
43.852
*
247.570
Adj SS
116.266
87.452
43.852
*
Adj MS
116.266
21.863
10.963
*
F
10.61
1.99
**
P
0.031
0.260
Analysis of Variance for MeSH, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
0.021563
0.174265
0.005715
*
0.201542
Adj SS
0.021563
0.174265
0.005715
*
Adj MS
0.021563
0.043566
0.001429
*
F
15.09
30.49
**
P
0.018
0.003
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
Good Vs Weak
Table 9 Anova of VSC Good Vs Weak
General Linear Model: DMS, H2S, MeSH versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
5
Values
good, weak
12M, 15M, 2M, 6M, 9M
Analysis of Variance for DMS, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
199.62
1332.04
65.53
*
1597.18
Adj SS
199.62
1332.04
65.53
*
Adj MS
199.62
333.01
16.38
*
F
12.18
20.33
**
P
0.025
0.006
Analysis of Variance for H2S, using Adjusted SS for Tests
Source
DF
Seq SS
Adj SS
Adj MS
F
P
221
treatment
time
treatment*time
Error
Total
1
4
4
0
9
31.2594
25.5177
8.9782
*
65.7552
31.2594
25.5177
8.9782
*
31.2594
6.3794
2.2445
*
13.93
2.84
**
0.020
0.168
Analysis of Variance for MeSH, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
0.046490
0.128501
0.026157
*
0.201148
Adj SS
0.046490
0.128501
0.026157
*
Adj MS
0.046490
0.032125
0.006539
*
F
7.11
4.91
**
P
0.056
0.076
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
TKN
Heat-shocked Vs Pasteurized
Table 10 Anova of TKN Heat-Shocked Vs Pasteurized
General Linear Model: WSN, TCA, PTA versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
10
Values
HS, P
10m, 12m, 14m, 16m, 18m, 20m, 2m, 4m, 6m, 8m
Analysis of Variance for WSN, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
9
9
0
19
Seq SS
22.4890
65.5854
1.5960
*
89.6704
Adj SS
22.4890
65.5854
1.5960
*
Adj MS
22.4890
7.2873
0.1773
*
F
126.82
41.09
**
P
0.000
0.000
Analysis of Variance for TCA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
9
9
0
19
Seq SS
0.7169
23.8577
0.6346
*
25.2092
Adj SS
0.7169
23.8577
0.6346
*
Adj MS
0.7169
2.6509
0.0705
*
F
10.17
37.59
**
P
0.011
0.000
Analysis of Variance for PTA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
9
9
0
19
Seq SS
0.04105
2.89366
0.01552
*
2.95023
Adj SS
0.04105
2.89366
0.01552
*
Adj MS
0.04105
0.32152
0.00172
*
F
23.80
186.44
**
P
0.001
0.000
222
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
TCCA Vs CRP
Table 11 TKN TCCA Vs CRP
General Linear Model: WSN, TCA, PTA versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
8
Values
CRP, TCCA
10m, 12m, 14m, 15m, 2m, 4m, 6m, 8m
Analysis of Variance for WSN, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
7
7
0
15
Seq SS
4.0102
132.8055
1.1492
*
137.9648
Adj SS
4.0102
132.8055
1.1492
*
Adj MS
4.0102
18.9722
0.1642
*
F
24.43
115.57
**
P
0.002
0.000
Analysis of Variance for TCA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
7
7
0
15
Seq SS
1.4495
16.4812
0.3755
*
18.3062
Adj SS
1.4495
16.4812
0.3755
*
Adj MS
1.4495
2.3545
0.0536
*
F
27.02
43.89
**
P
0.001
0.000
Analysis of Variance for PTA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
7
7
0
15
Seq SS
0.72285
4.10058
0.12258
*
4.94600
Adj SS
0.72285
4.10058
0.12258
*
Adj MS
0.72285
0.58580
0.01751
*
F
41.28
33.45
**
P
0.000
0.000
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
Adjunct culture Vs Not
Table 12 TKN Adjunct culture Vs Not
General Linear Model: WSN, TCA, PTA versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
5
Values
Adj, Without
10m, 2m, 4m, 6m, 8m
223
Analysis of Variance for WSN, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
5.114
83.168
1.857
*
90.139
Adj SS
5.114
83.168
1.857
*
Adj MS
5.114
20.792
0.464
*
F
11.01
44.78
**
P
0.029
0.001
Analysis of Variance for TCA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
1.19657
5.39730
0.36319
*
6.95706
Adj SS
1.19657
5.39730
0.36319
*
Adj MS
1.19657
1.34933
0.09080
*
F
13.18
14.86
**
P
0.022
0.011
Analysis of Variance for PTA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
0.11281
1.25264
0.08796
*
1.45342
Adj SS
0.11281
1.25264
0.08796
*
Adj MS
0.11281
0.31316
0.02199
*
F
5.13
14.24
**
P
0.086
0.012
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
Good Vs Weak
Table 13 TKN Good Vs Weak
General Linear Model: WSN, TCA, PTA versus treatment, time
Factor
treatment
time
Type
random
fixed
Levels
2
5
Values
Good, Weak
12m, 15m, 2m, 6m, 9m
Analysis of Variance for WSN, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
26.1002
46.5563
2.8000
*
75.4564
Adj SS
26.1002
46.5563
2.8000
*
Adj MS
26.1002
11.6391
0.7000
*
F
37.29
16.63
**
P
0.004
0.009
Analysis of Variance for TCA, using Adjusted SS for Tests
Source
treatment
time
DF
1
4
Seq SS
2.45150
5.36704
Adj SS
2.45150
5.36704
Adj MS
2.45150
1.34176
F
118.81
65.03
P
0.000
0.001
224
treatment*time
Error
Total
4
0
9
0.08253
*
7.90107
0.08253
*
0.02063
*
**
Analysis of Variance for PTA, using Adjusted SS for Tests
Source
treatment
time
treatment*time
Error
Total
DF
1
4
4
0
9
Seq SS
2.14584
5.83349
0.15910
*
8.13843
Adj SS
2.14584
5.83349
0.15910
*
Adj MS
2.14584
1.45837
0.03978
*
F
53.95
36.66
**
P
0.002
0.002
** Denominator of F-test is zero or undefined.
S = *
* NOTE * Could not graph the specified residual type because MSE = 0 or the
degrees of freedom for error = 0.
225
APENDIX B
PCA CORRELATION MATRIX
HEAT SHCOKED VS PASTEURIZATION
Table 14 Correlation matrix for HEAT SHCOKED VS PASTEURIZATION
Principal Component Analysis:
Eigenanalysis of the Correlation Matrix
Eigenvalue
Proportion
Cumulative
6.5409
0.503
0.503
2.9590
0.228
0.731
1.2323
0.095
0.826
0.7095
0.055
0.880
0.5463
0.042
0.922
Eigenvalue
Proportion
Cumulative
0.0950
0.007
0.988
0.0674
0.005
0.993
0.0522
0.004
0.997
0.0324
0.002
0.999
0.0100
0.001
1.000
Variable
PC1
Temperature 0.360
Time
0.072
seg 12-20
0.325
seg 20-25
0.258
seg 25-30 -0.291
seg 30-35 -0.285
seg 35-40
0.331
seg 40-45
0.299
seg 45-50 -0.150
seg 50-55 -0.290
seg 55-60 -0.346
seg 60-65 -0.183
PC2
0.182
-0.440
0.074
0.144
-0.193
-0.292
-0.104
-0.225
0.498
0.397
0.123
-0.353
PC3
0.077
0.500
0.221
-0.467
0.003
0.182
-0.024
-0.310
0.198
-0.170
0.080
-0.471
PC4
-0.093
-0.161
0.264
0.234
0.320
0.176
-0.379
-0.260
-0.409
0.137
-0.237
-0.053
PC5
-0.364
0.280
0.455
0.208
-0.228
-0.404
-0.171
-0.298
-0.162
0.036
0.079
0.025
Variable
PC9
Temperature 0.052
Time
0.076
seg 12-20 -0.058
seg 20-25
0.549
seg 25-30 -0.260
seg 30-35 -0.297
seg 35-40
0.315
seg 40-45 -0.260
seg 45-50
0.055
seg 50-55 -0.074
seg 55-60
0.591
seg 60-65 -0.071
PC10
0.322
-0.002
0.633
0.081
0.181
-0.085
-0.030
0.373
0.004
0.287
0.230
0.292
PC11
0.662
0.381
-0.272
0.117
-0.036
-0.117
-0.296
-0.005
0.291
0.033
-0.146
0.176
PC12
-0.304
0.271
0.095
0.038
0.629
0.112
-0.119
-0.212
0.530
-0.189
0.198
0.027
PC13
-0.164
0.432
-0.261
0.440
-0.075
0.396
-0.093
0.309
-0.141
0.040
-0.136
-0.019
TCCA Vs CRP
0.3552
0.027
0.949
PC6
0.149
-0.035
0.052
-0.198
-0.670
0.314
-0.259
-0.172
0.066
0.071
-0.322
0.419
0.2357
0.018
0.968
PC7
-0.034
-0.185
-0.064
0.196
-0.314
0.164
-0.397
0.168
-0.119
-0.505
0.547
-0.188
0.1641
0.013
0.980
PC8
-0.063
0.016
0.029
-0.040
-0.223
0.001
0.520
-0.217
-0.289
0.246
0.521
0.241
226
Table 15 Coreelation matrix of TCCA Vs CRP
Principal Component Analysis:
Eigenanalysis of the Correlation Matrix
35 cases used, 1 cases contain missing values
Eigenvalue
Proportion
Cumulative
7.4190
0.571
0.571
1.7891
0.138
0.708
1.2763
0.098
0.806
0.8856
0.068
0.875
0.4454
0.034
0.909
Eigenvalue
Proportion
Cumulative
0.1580
0.012
0.977
0.1165
0.009
0.986
0.0801
0.006
0.992
0.0552
0.004
0.997
0.0426
0.003
1.000
Variable
Location
Time
Batch
seg 12-20
seg 20-25
seg 25-30
seg 30-35
seg 35-40
seg 40-45
seg 45-50
seg 50-55
seg 55-60
seg 60-65
PC1
0.060
0.334
0.269
0.255
0.317
-0.307
-0.335
0.225
-0.122
-0.289
-0.307
-0.326
-0.266
PC2
0.573
0.043
-0.213
-0.409
-0.196
-0.242
0.001
0.359
0.434
-0.183
-0.242
-0.062
0.061
PC3
0.524
-0.208
0.294
-0.268
0.091
-0.067
0.020
-0.323
-0.557
-0.219
-0.067
0.091
-0.071
PC4
0.006
-0.173
-0.352
0.258
0.019
0.182
-0.080
0.357
-0.363
-0.451
0.182
0.084
0.517
PC5
-0.034
-0.059
0.178
0.027
-0.404
0.385
-0.118
0.398
-0.070
-0.166
0.385
0.347
-0.571
Variable
Location
Time
Batch
seg 12-20
seg 20-25
seg 25-30
seg 30-35
seg 35-40
seg 40-45
seg 45-50
seg 50-55
seg 55-60
seg 60-65
PC9
0.288
0.178
0.245
0.174
0.484
0.047
-0.176
-0.006
-0.096
-0.228
0.527
0.427
-0.076
PC10
-0.198
0.131
-0.090
-0.052
0.187
-0.264
0.332
0.079
-0.227
-0.019
-0.351
0.563
0.469
PC11
-0.060
0.140
-0.081
-0.224
-0.365
-0.068
0.572
-0.377
-0.002
-0.288
0.316
0.170
-0.318
PC12
0.089
-0.369
0.441
0.590
-0.192
0.127
0.377
-0.042
0.018
0.302
-0.025
0.153
-0.002
PC13
0.392
-0.607
-0.293
-0.110
0.205
0.285
0.163
-0.091
0.017
-0.424
-0.144
-0.016
0.147
0.2948
0.023
0.932
PC6
-0.125
0.187
0.656
-0.085
-0.206
-0.070
0.417
0.340
-0.072
0.035
-0.070
-0.054
0.407
0.2420
0.019
0.950
PC7
0.148
0.344
0.191
-0.068
0.131
0.196
-0.371
-0.175
0.156
0.083
0.196
0.610
0.317
0.1953
0.015
0.965
PC8
0.000
0.290
-0.008
0.415
-0.384
-0.310
0.063
-0.355
0.183
-0.436
-0.310
0.175
0.019
ADJUCNT CULTURE Vs NOT
Table 16 Correlation matrix of Adj Vs Not
Principal Component Analysis:
Eigenanalysis of the Correlation Matrix
35 cases used, 1 cases contain missing values
Eigenvalue
7.6670
2.0734
1.1046
0.5703
0.4492
0.3360
0.2211
0.1501
227
Proportion
Cumulative
0.590
0.590
0.159
0.749
0.085
0.834
0.044
0.878
0.035
0.913
Eigenvalue
Proportion
Cumulative
0.1360
0.010
0.978
0.1142
0.009
0.986
0.0863
0.007
0.993
0.0581
0.004
0.997
0.0335
0.003
1.000
Variable
Time
Culture
Batch
seg 12-20
seg 20-25
seg 25-30
seg 30-35
seg 35-40
seg 40-45
seg 45-50
seg 50-55
seg 55-60
seg 60-65
PC1
0.294
0.049
-0.146
0.220
0.256
-0.293
-0.328
0.329
0.320
-0.300
-0.302
-0.314
0.304
PC2
-0.294
0.254
0.570
0.358
0.274
-0.227
-0.055
0.153
0.166
0.235
0.218
-0.149
-0.299
PC3
-0.129
-0.860
0.223
0.050
0.171
0.230
-0.029
0.144
0.159
-0.151
0.186
-0.032
0.062
PC4
-0.270
0.041
-0.146
0.737
-0.028
0.369
-0.006
0.031
-0.213
-0.039
-0.323
0.244
0.108
PC5
0.254
0.002
0.358
0.224
-0.787
-0.063
0.081
0.073
0.236
-0.243
0.062
0.060
0.009
Variable
Time
Culture
Batch
seg 12-20
seg 20-25
seg 25-30
seg 30-35
seg 35-40
seg 40-45
seg 45-50
seg 50-55
seg 55-60
seg 60-65
PC9
-0.089
0.002
-0.182
-0.014
0.017
-0.331
-0.058
0.002
-0.225
-0.126
-0.713
0.140
0.501
PC10
0.342
-0.044
0.009
0.145
0.058
-0.074
-0.775
-0.017
-0.248
-0.171
0.007
-0.384
-0.111
PC11
0.034
-0.084
-0.268
-0.678
0.019
0.400
-0.045
0.059
0.201
-0.473
-0.067
-0.149
0.066
PC12
0.288
-0.085
0.154
-0.157
0.029
-0.301
0.284
-0.360
-0.445
-0.474
0.212
0.262
-0.152
PC13
0.434
-0.657
-0.034
0.375
0.112
0.330
0.234
-0.035
0.102
-0.017
-0.134
-0.027
0.177
0.026
0.939
PC6
0.084
0.191
-0.070
-0.051
0.108
0.171
-0.478
0.173
-0.113
-0.488
0.391
0.378
-0.321
0.017
0.956
PC7
-0.029
0.101
0.201
-0.261
0.034
0.217
-0.248
0.111
0.476
0.346
-0.258
0.540
0.233
0.012
0.967
PC8
0.437
0.120
0.106
0.148
0.425
0.129
0.578
-0.241
0.237
-0.230
0.069
0.243
-0.005
GOOD VS WEAK
Table 17 Correlation matrix Good Vs Weak
Principal Component Analysis:
Eigenanalysis of the Correlation Matrix
Eigenvalue
Proportion
Cumulative
5.2975
0.407
0.407
4.0625
0.313
0.720
1.1362
0.087
0.807
0.6540
0.050
0.858
0.5932
0.046
0.903
Eigenvalue
Proportion
Cumulative
0.0988
0.008
0.986
0.0925
0.007
0.993
0.0421
0.003
0.996
0.0287
0.002
0.999
0.0172
0.001
1.000
PC1
PC2
PC3
PC4
PC5
Variable
0.4589
0.035
0.939
0.3741
0.029
0.967
0.1443
0.011
0.979
PC6
PC7
PC8
228
Quality
Time
Batch
seg 12-20
seg 20-25
seg 25-30
seg 30-35
seg 35-40
seg 40-45
seg 45-50
seg 50-55
seg 55-60
seg 60-65
0.000
0.383
0.238
0.339
0.390
-0.182
-0.374
0.295
0.293
-0.210
-0.312
-0.198
0.061
0.472
0.006
0.006
0.073
0.056
-0.466
-0.072
-0.386
-0.072
-0.080
-0.191
0.389
-0.445
0.339
-0.182
0.432
0.031
0.066
0.150
0.075
0.219
-0.389
-0.517
0.152
0.225
0.304
0.358
0.073
-0.313
0.289
-0.182
0.339
-0.052
0.062
0.192
0.247
-0.303
0.411
0.414
0.082
0.296
-0.103
-0.613
-0.049
0.224
-0.517
-0.052
-0.348
-0.045
-0.265
-0.051
0.010
Variable
Treatment
Time
Batch
seg 12-20
seg 20-25
seg 25-30
seg 30-35
seg 35-40
seg 40-45
seg 45-50
seg 50-55
seg 55-60
seg 60-65
PC9
-0.042
-0.417
0.181
0.383
0.047
0.271
0.251
-0.158
-0.436
0.370
0.250
0.267
-0.143
PC10
0.167
0.120
0.247
0.080
-0.139
-0.162
-0.576
0.153
-0.146
0.357
0.064
-0.306
-0.493
PC11
-0.046
-0.081
0.598
-0.039
0.168
-0.603
0.197
-0.099
-0.268
-0.295
0.030
-0.146
0.106
PC12
-0.339
0.668
0.144
0.007
-0.070
-0.165
0.064
-0.082
0.061
0.209
0.214
0.526
-0.071
PC13
0.552
0.267
-0.027
0.057
-0.036
-0.066
0.417
-0.028
0.155
0.326
0.384
-0.330
0.229
0.022
0.108
-0.443
0.217
-0.138
0.005
-0.441
0.053
0.060
-0.322
0.636
0.077
-0.100
-0.080
0.032
0.542
0.124
-0.213
0.501
-0.227
-0.427
0.298
0.000
0.088
0.070
-0.226
0.263
-0.320
-0.015
-0.539
0.166
-0.006
-0.028
0.175
0.672
-0.120
0.088
0.067
0.000
Download